1. Introduction
In research of legal argumentation different aspects of the process of legal justification have been the object of study. Some researchers consider legal justification as a rational activity and for this reason are interested in the rules that should be observed in rational legal discussions. Others consider legal justification as a rhetorical practice and are interested in the way in which judges operate in steering the discussion in the direction that is desirable from the perspective of certain legal goals.

That both aspects of the legal ‘enterprise’, rational dispute resolution and a rhetorical orientation to a particular result through strategic manoeuvring, can also be reconciled is something that has received little attention in research of legal argumentation. The aim of this contribution is to analyse the way in which courts try to reconcile the dialectical goal of resolving the difference of opinion in a rational way with the rhetorical goal of steering the discussion in a particular direction that is desirable from the perspective of a particular development of law.

To this end I shall analyse the strategic manoeuvring in the justification of the Dutch Supreme Court in the famous case of the ‘Unworthy Spouse’ in which a spouse who had murdered his wife claimed his share in the matrimonial community of property. In this case it had to be established whether and on what grounds an exception to article 1:100 of the Dutch Civil Code, that entitles a spouse to his share in the community of property, can be justified. (For an overview of the relevant legal rules see A at the end of this contribution.) The District Court, the Court of Appeal and the Supreme Court all agreed that an exception should be made and they all justified the exception by referring to certain legal principles that can be summarized as ‘crime does not pay’.[i] However, with regard to the exact argumentative role of the legal principles the Supreme Court adopts another position than the other courts but it does not express this position explicitly but presents it in an indirect way as the interpretation of the decision of the Court of Appeal, thereby giving another interpretation of the argumentative role of the legal principles than was originally intended by the Court of Appeal.

In my contribution I shall describe how the Dutch Supreme Court manoeuvres strategically in its role as court of cassation when attributing a different argumentative role to the legal principles than is intended by the Court of Appeal.[ii] I shall explain how the Supreme Court operates strategically in its capacity of court of cassation to promote a particular development of law with respect to the role of legal principles to make an exception to a rule of law.

The central question in the case of the Unworthy Spouse is whether behaviour that can be considered ‘unacceptable from the perspective of a sense of justice’ or ‘repugnant to justice’ must also be considered as unacceptable from the perspective of civil law when there are no existing rules on the basis of which this behaviour can be characterized as unacceptable. In this case the question is whether a spouse (in this case L.) who has murdered his 72 year old wife (mrs. Van Wylick) after 5 weeks of marriage and who has been convicted of murder in a criminal procedure, still has a right to his legal share in the marital community property on the basis of article 1:100 clause 1 (old) of the Dutch Civil Code, and if he does not have such a right how the exception should be justified for this case.

In this case the Court of Appeal decides that L. Does not have a right to his legal share in the marital community of property, making an exception to the rule of 1:00 of the Civil code for this case. The Court of Appeal justifies the exception by referring to two legal principles. The first principle is that he, who deliberately causes the death of someone else, who has benefited and favoured him, should not profit from this favour (P1). The second principle is that one should not profit from the deliberately caused death of someone else (P2).  Furthermore the Court of Appeal argues as an ‘obiter dictum’ that also the requirements of reasonableness and fairness would justify making an exception in this particular case. An overview of the main structure of the argumentation of the Court of Appeal is given in scheme 1A.

The Supreme Court also answers this question positively. However, the Supreme Court gives another justification of the exception by considering the exception on the basis of reasonableness and fairness as the main argument. An overview of the main structure of the argumentation of the Supreme Court is given in scheme 1B.

As is indicated in scheme IB, in support of this main argument (1.1.1), the Supreme Court mentions the two legal principles in 1.1.2.1b in combination with the exceptional circumstances of this case. In doing so, the Supreme Court departs from the way in which the argument of reasonableness and fairness was presented by the Court of Appeal, i.e. as an obiter dictum (argument 1.1.2), while the two legal principles were presented by the Court of Appeal as the independent main argument 1.1.1.

As is mentioned by the annotator, from the perspective of legal certainty the Supreme Court wants to give a signal to the legal community that general legal principles cannot constitute a reason for making an exception to a legal rule that forms one of the cornerstones of Dutch family law. For this reason the Supreme Court chooses for the ‘safe’ option of restricting the exception to the concrete case by using the derogating function of reasonableness and fairness (which will be introduced in the new article 6:2 of the Civil Code) as the main argumentation 1.1.1 and the legal principles as supporting coordinative argumentation (1.1.2.1b) in combination with the exceptional circumstances (1.1.2.1a).

In this paper I will answer the question what the discussion strategy of the Supreme Court in rejecting the cassation grounds and in changing the argumentative role of the legal principles exactly amounts to from the perspective of the space he has to manoeuvre strategically as a court of cassation. In my analysis of the argumentation strategy of the Supreme Court I use the concept of strategic manoeuvring developed by van Eemeren (2010) and van Eemeren and Houtlosser (2006, 2007). In their approach strategic manoeuvring is conceived as an attempt to reconcile the dialectical goal of resolving a difference of opinion in a reasonable way with the rhetorical goal of steering the resolution in a particular direction.

Van Eemeren and Houtlosser describe a discussion strategy as a methodical design of discussion moves aimed at influencing the result of a particular discussion stage, and the discussion as a whole, in the desired direction. A discussion strategy consists of a systematic, co-ordinated and simultaneous exploitation of the options available in a particular stage of the discussion.

Starting from this conception I shall show that the discussion strategy of the Supreme Court can be described as a consistent effort in the different stages of a critical discussion to steer the discussion in the desired direction.[iii] I characterize the choices the Supreme Court makes in the different stages as a methodical design to steer the outcome of the discussion in the preferred direction, within the boundaries created by the institutional conventions for the discussion in cassation.

2. Analysis of the discussion strategy of the Supreme Court in the case of the ‘Unworthy  Spouse’
The aim of the procedure in cassation in the Netherlands is to establish what the law in a particular case should be and how the law should be applied in that case. To this end, in this case the Supreme Court must decide whether the decision of the Court of Appeal is in accordance with the law. For this case this implies that the Supreme Court must investigate whether the rules of law that are applied by the Court of Appeal have been applied correctly.

From this perspective, the dialectical goal of the discussion is to establish whether the protagonist in the case in cassation, the Court of Appeal, has defended its decision successfully against the attacks of the antagonist, the plaintiff in cassation, in light of the common starting points, the rules of law, so that the Court of Appeal can maintain his standpoint, or whether it has been attacked successfully. In this case the Supreme Court tries to reconcile this dialectical goal with the rhetorical goal to steer the discussion in the desired direction, i.e. to convince the audience that application of the rule without making an exception for the concrete case would be unacceptable from the perspective of justice.[iv] To attain this rhetorical goal, the Supreme Court gives a particular interpretation of the system of the law of inheritance by attaching a particular argumentative role to the general legal principles as a legal ground for making an exception to article 1:100 clause 1 of the Civil Code.

To be able to decide that the decision of the Court of Appeal can be maintained, the Supreme Court adopts a particular discussion strategy that consists of a combination of two ‘moves’. First, the Supreme Court wants to be able to decide in the concluding stage of the discussion that the attacks of the plaintiff in cassation L have failed. To realize this aim, in the argumentation stage the Supreme Court must decide that the argumentation of the Court of Appeal is in accordance with the common starting points. To be able to decide this, in the opening stage the Supreme Court must select those starting points that make this evaluation of the argumentation of the Court of Appeal possible.

Second, the Supreme Court wants to give a decision that makes clear that an exception to the rules of family law and the law of inheritance can only be made in very special circumstances. For this reason the Supreme Court must select those starting points that are desirable in light of this view on the development of these branches of law. For this reason, in the opening stage the Supreme Court does not only decide about the role of reasonableness and fairness and certain legal principles as starting points, but also about their argumentative role.

In my analysis I shall explain how this discussion strategy manifests itself in the justification of the decision of the Supreme Court as given in scheme 1B.[v] I shall do this on the basis of the statements of the Supreme Court in the legal considerations 3.2-3.5 (see F at the end of this contribution) that I shall analyse in terms of certain moves in a critical discussion.

The confrontation stage
In this case, the confrontation stage that is intended at realizing the dialectical goal of establishing the difference of opinion, is represented by the cassation grounds formulated by the plaintiff in which he formulates his objections against the decision of the Court of Appeal.[vi]  The plaintiff is of the opinion that the Court of Appeal has made a mistake in applying the law by deciding erroneously that certain legal principles apply and by deciding erroneously that it is justified to make an exception to article 1:100 clause 1 of the Civil Code on the basis of reasonableness and fairness. Because the plaintiff determines the content and scope of the difference of opinion, the Supreme Court has no space to manoeuvre strategically in this discussion stage.

The opening stage
In the opening stage the discussion strategy consists of a methodical design of discussion moves aimed at reconciling the dialectical goal of establishing the common starting points with the rhetorical goal of establishing those starting points that are advantageous in view of his final goal of dismissing the appeal in cassation so that the decision of the Court of Appeal can be maintained as well as a particular development of law. The Supreme Court exploits the space he has on the basis of his dialectical role to establish the common legal starting points in a specific way.

In civil procedure in the Netherlands the latitude to establish common legal starting points is specified in article 48 of the Code of Civil Procedure that gives the judge, in this case the Supreme Court, the authority to formulate the legal grounds. In this case it uses this latitude to formulate the legal grounds on the basis of which the exception to article 1:100 clause of the Civil Code can be justified.

The discussion strategy in the opening stage amounts to the following. The Supreme Court chooses those starting points from the topical potential that it needs to steer the result of the opening stage in the desired direction: it chooses those starting points that it needs in the argumentation stage to be able to evaluate the attack of the plaintiff as a failed attack on the argumentation of the Court of Appeal. In doing so the Supreme Court tries to adapt to the preferences of the legal community by taking into account that acknowledging the claim of the plaintiff would be ‘unacceptable for the sense of justice’, as is also stressed by the Advocate-General Langemeijer.

In the old matrimonial property law there was not a rule specifying when someone is unworthy to inherit. To avoid a result that would be unacceptable to the sense of justice therefore the Supreme Court must create a possibility to make an exception to article 1:100 clause 1 of the Civil Code on the basis of certain common legal starting points. The Supreme Court establishes the common starting points by acknowledging that it is possible to make an exception to article 1:100 and it establishes that this exception can be justified on the basis of reasonableness and fairness and on the basis of certain legal principles. In doing so the Supreme Court rebuts the statement of the plaintiff that the exception can not be justified in this way.

Apart from this decision about the status of reasonableness and fairness and the legal principles as common legal starting points, the Supreme Court also decides about the argumentative function of these common starting points. The Supreme Court does this in an implicit way with the following statement in consideration in which it rejects the statements in the cassation grounds of the plaintiff:

‘As appears from the cited formulation, in this context the legal principles only play the role that they have contributed to the decision of the court that the requirements of reasonableness and fairness make the exertion of his right to his share in the inheritance inadmissible. As far as the parts A and B read in legal consideration 5.18 that the court has used these principles as a direct legal ground for denying this right, they lack a factual basis’.[vii] As is shown in the analysis of the argumentation of the Court of Appeal in scheme 1A and the analysis of the argumentation of the Supreme Court in scheme 1B, the Supreme Court gives an interpretation of the argumentation of the Court of Appeal that departs from the way in which the court has intended it. The Supreme Court gives the legal principles the function of subordinate argumentation and does not consider them as independent argumentation as they were presented by the Court of Appeal.

The argumentation stage
In the argumentation stage the discussion strategy consists of a methodical design of discussion moves aimed at giving a positive evaluation of the argumentation of the Court of Appeal in light of the attacks by the plaintiff. In the argumentation stage the Supreme Court tries to reconcile the dialectical goal of establishing the acceptability of the argumentation of the Court of Appeal on the basis of common testing methods in light of the attacks of the plaintiff with the rhetorical goal of evaluating the attacks of the plaintiff in such a way that these attacks fail. To attain this, the Supreme Court uses the common starting points formulated in the opening stage. In doing so, the Supreme Court exploits the space it has within his dialectical task and the authority it has on the basis of the legal rules to evaluate the argumentation in a special way.

The discussion strategy manifests itself first in the statements in the decision in which the Supreme Court decides in legal consideration 3.2 that the grounds of cassation A and B ‘cannot lead to cassation’ because they ‘lack interest’, ‘lack a factual basis’ and ‘depart from a wrong conception of the law’. The strategy manifests itself second in the decision in legal consideration 3.3 cited above that the statement about the exception on the basis of reasonableness and fairness from part C is wrong.

These decisions imply that the attack of the plaintiff (in cassation grounds A and B) on argumentation line 1.1 of the Court of Appeal has failed because the legal principles do exist. The attack (in cassation ground C) on argumentation line 1.2 also fails because the Supreme Court  decides that the possibility to make an exception is possible, but only  in very special circumstances.

To be able to make the choice from the topical potential that is most suitable to reach the desired result of the argumentation stage, the Supreme Court has prepared these choices in the opening stage. The Supreme Court chooses to present part C of the cassation grounds as a failing attempt to attack the decision by using the formulation that says that C ‘contests in vain’ part 5.18 of the argumentation. The Supreme Court presents the attacks in the cassation grounds A and B as failing attacks and characterizes them in legal terms as attacks that cannot lead to cassation ‘because of lack of interest’.

The concluding stage
Finally, in the concluding stage, the Supreme Court decides on the basis of this evaluation of the grounds of cassation in the argumentation stage that the appeal in cassation must be dismissed, which implies that the decision of the Court of Appeal can remain intact. The Supreme Court uses the space he has within his dialectical tasks and the authority he has on the basis of the applicable legal rules to present the choices he has made in the previous stages as a justification of his final decision.

The discussion strategy of the Supreme Court implies that it does two things at the same time. First it decides that the attacks by the plaintiff on the argumentation of the Court of Appeal have failed so that the decision can remain intact. Second, the Supreme Court gives an implicit interpretation of the argumentation of the Court of Appeal that departs from the way in which the argumentation was intended. This discussion move is not necessary to accomplish the dialectical goal of establishing the acceptability of the argumentation of the Court of Appeal because the Supreme Court can dismiss the appeal without this interpretation. The differing interpretation can be considered as an implicit ‘obiter dictum’ that the Supreme Court gives as a signal to the legal community in his capacity as judge of cassation to point out how the law should be developed. By choosing an interpretation in which the Supreme Court justifies the exception to article 1.100 clause 1 of the Civil Code on the basis of reasonableness and fairness that is supported by an appeal to the legal principles instead of a direct appeal to the legal principles, the Supreme Court makes indirectly clear that it does not want to consider the legal principles as the main argument and therefore as the main reason to make an exception.

3. Conclusion
With this analysis of the discussion strategy of the Supreme Court to establish the legal and argumentative function of certain legal principles in a concrete case as a systematic effort in the various discussion stages I have clarified how the Supreme Court combines a rational resolution of legal disputes and a rhetorical choice and presentation of discussion moves. The Supreme Court uses the space it has within the boundaries of his dialectical role and the applicable institutional rules to manoeuvre strategically to resolve the difference of opinion and at the same time establish the argumentative role of the applicable legal principles. In the opening stage the Supreme Court uses the space it has within the institutional boundaries to establish the common legal starting points. It establishes the content of the common legal starting points in such a way that it is able to give a negative evaluation of the attacks of the plaintiff in the argumentation stage. On the basis of this negative evaluation it can finally dismiss the appeal in the concluding stage. At the same time, the Supreme Court also uses the space it has within the institutional boundaries to establish the argumentative role of the common legal starting points. The Supreme Court decides that in making an exception to rule 1:100 of the law of inheritance, this exception must be restricted to the concrete case.

NOTES
[i]  See the decisions published in NJ 1988/992, 8-4-1987, NJ 1989/369, 24-11-1988, NJ 1991/593, 7-12-1990.
[ii] Cf. the case of Riggs v. Palmer, 115 N.Y. 506, 22 N.E. 188 (1889) mentioned by Dworkin (1986, pp. 15-20) as an example of a systematic interpretation of the law of inheritance with the aim of clarifying the underlying principles.
[iii] For other analyses of the strategic manoeuvring in legal decisions see Feteris (2008, 2009a and 2009b).
[iv] In the case of legal justification the audience of the Dutch Supreme Court is a composite audience consisting of various ‘groups’. Firstly the audience consists of the parties in dispute. Secondly, in cases of appeal and cassation, the audience also consists of the judges that have taken prior decisions. Thirdly, the audience consists of members of the legal community of legal practitioners such as other judges and lawyers for whom the justification provides information about the way in which the law needs to be applied according to the Supreme Court. Although the decisions do not have the status of precedents, other judges and lawyers take into account the opinions of the Supreme Court in similar cases.
[v] See for a more extended analysis of the decision of the Supreme Court analysis D at the end of this contribution.
[vi] For the relevant parts of the decision of the Court of Appeal see E at the end of this contribution. For a more extended analysis of the argumentation of the Court of Appeal see B at the end of this contribution. For an analysis of the argumentation of the plaintiff see C at the end of this contribution)
[vii] See for the complete text of the justification of the Supreme Court F at the end of this contribution.

REFERENCES
Dworkin, R. (1986). Law’s empire. London: Fontana.
Eemeren, F.H. van (2010). Strategic manoeuvering in argumentative discourse. Extending the pragma-dialectical theory of argumentation. Amsterdam: John Benjamins.
Eemeren, F.H. van, & Houtlosser, P. (2006). Strategic maneuvering: A synthetic recapitulation. Argumentation, 20, 4, p. 377-380.
Eemeren, F.H. van, & Houtlosser, P. (2007). Seizing the occasion: Parameters for analysing ways of strategic manoeuvring. In: F.H. van Eemeren, J.A. Blair, Ch.A. Willard & B. Garssen (eds.), Proceedings of the sixth conference of the International Society for the Study of Argumentation. Amsterdam: SicSat, p. 375-381.
Feteris, E.T. (2008). Strategic maneuvering with the intention of the legislator in the justification of judicial decisions’. Argumentation. Vol. 22, pp. 335-353.
Feteris, E.T. (2009a). ‘Strategic manoeuvring in the justification of judicial decisions’. In: F. H. van Eemeren (ed.), Examining argumentation in context. Fifteen studies on strategic manoeuvering. Amsterdam: John Benjamins, p. 93-114.
Feteris, E.T. (2009b) ‘Strategic manoeuvring with linguistic arguments in the justification of legal decisions’. Proceedings of the Second Conference Rhetoric in Society, Leiden University, 22-23 January 2009. (cd-rom).

Appendix
A. Legal rules applied in the case of the Unworty Spouse
Article 1:100 of the Old Dutch Civil Code
1. The spouses have an equal share in this divided community of property, unless a different division is established by means of a marriage settlement (…).
Article 4.3 of the New Dutch Civil Code
1.Legally unworthy to profit from an inheritance are: He who has been condemned irrevocably because he has killed the deceased, he who has tried to kill the deceased or he who has prepared to kill the deceased or has participated in preparing to kill the deceased.
Article 6:248, 2 of the Dutch Civil Code

An arrangement that is valid between the creditor and the debtor on the basis of the law, a custom or a legal act, does not apply if this is unacceptable from the perspective of the standards of reasonableness and fairness

Article 3:12 of the Dutch Civil Code
When establishing what reasonableness and fairness require, generally accepted legal principles, legal convictions that are generally accepted in the Netherlands, and social and personal interests in a particular case, should be taken into account.
B. Decision of the Court of appeal
1. The claim of L, stating that he is entitled to his share in the marital community of property, must be dismissed
1.1 L. should not profit from the marital community of property (5.17, 5.18)
1.1.1 In the special circumstances of the concrete case an exception to the legal division on the basis of article 1:100 of the Dutch Civil Code is justified on the basis of the following two legal principles:
1.1.1.1a He, who deliberately causes the death of someone else, who has benefited favoured him, should not profit from this favour (5.13) (legal principle P1)
1.1.1.1a.1 Article 3:959 of the Dutch Civil Code and article 4:1725 sub 2e of the Dutch Civil Code (5.14)
1.1.1.1b One should not profit from the deliberately caused death of someone else (legal principle P2)
1.1.1.1b.1 Article 3:885 sub 1e of the Dutch Civil Code
1.1.2 In the concrete case an exception to the legal division of the marital community of property on the basis of article 1:100 of the Dutch Civil Code is justified on the basis of reasonableness and fairness as specified in article 6:2 section 2 of the New Dutch Civil Code
1.1.2.1a The exceptional circumstances of the concrete case
1.1.2.1b He, who deliberately causes the death of someone else, who has favoured him, should not profit from this favour (5.13) (legal principle P1)
1.1.2.1b.1 Article 3:959 of the Dutch Civil Code and section 4:1725 sub 2e of the Dutch Civil Code (5.14)
1.1.2.1c One should not profit from the deliberately caused death of someone else (legal principle P2)
1.1.2.1c.1 Article 3:885 sub 1e of the Dutch Civil Code

C. Argumentation of the plaintiff in cassation
1. The decision by the court in which it denies my claim that I am entitled to my share in the marital community of property must be nullified because the court has made mistakes in the application of the law
1.1a The court erroneously has based its decision on the two general legal principles P1 and P 2 (grounds of cassation A and B attacking argument 1.1.1)
1.1a.1a These principles do not exist
1.1a.1b These principles do not apply because I am not favoured by the marriage
1.1a.1b.1 The marital community op property is not a favour and I have not profited from the death of mrs. Van Wylick because I had already become the owner of half of the marital community on the basis of my marriage with her
1.1b On the basis of article 11 AB the judge is not allowed to make an exception to a clear legal rule on the basis of reasonableness and fairness (ground of cassation C attacking argument 1.1.2)

D. Decision of the Supreme Court
1 The claim of L, stating that he is entitled to his share the marital community of property, must be dismissed
1.1.1 In the concrete case an exception to the legal division of the marital community of property on the basis of article 1:100 of the Dutch Civil Code is justified on the basis of reasonableness and fairness as specified in clause 6:2 section 2 of the New Dutch Civil Code
1.1.1.1a The exceptional circumstances of the concrete case
1.1.1.1b In the concrete case an exception to the legal division on the basis of article 1:100 of the Dutch Civil Code is justified on the basis of the following two legal principles:
1.1.1.1b.1a He, who deliberately causes the death of someone else, who has favoured him, should not profit from this favour (5.13) (legal principle P1)
1.1.1.1b.1a.1 Article 3:959 of the Dutch Civil Code and article 4:1725 sub 2e of the Dutch Civil Code (5.14)
1.1.1.1b.1b One should not profit from the deliberately caused death of someone else (legal principle P2)
1.1.1.1b.1b.1 Article 3:885 sub 1e of the Dutch Civil Code

E. Text of the decision of the court of appeal NJ 1989/369, 24-11-1988
5.13 Since the district court has assumed that Mrs. Van Wylick intended with the marriage – that also according to L was a marriage of convenience- a financial benefit for L, the district court has rightly stressed that to the factual situation described in the foregoing the general legal principle is applicable that he, who has deliberately caused the death of someone else, who has favoured him, should not profit from the this favour.
(…)
5.16 In this context it is also important to mention that the aforementioned legal principle is closely related to another legal principle, i.e. that one should not profit form the deliberately caused death of someone else, which principle has among others been expressed in article 885 under 1 book 3 CC.
(…)
5.17 Application of the mentioned legal principles leads under the aforementioned facts and circumstances to the conclusion that L is not entitled to the benefit that is the consequence of the community of property created by the marriage without a marriage settlement (‘huwelijkse voorwaarden’) with mrs. van Wylick.
5.18 Also an examination of the claims of L in light of the requirements of reasonableness and fairness according to which he is supposed to behave in the community of property that is created by the marriage, as is stated by Brouwers c.s., leads to the conclusion that L should not profit from the marital community of property. In this case the court applies a strict standard because the appeal to reasonableness and fairness is aimed at preventing the claims of L completely. Also when applying such a strict standard the court is of the opinion that the claims of L must be considered as so unreasonable and unfair, in the aforementioned special circumstances of this case and also considered in light of the mentioned general legal principles, that the exertion of the claimed rights must be denied to him completely.

F. Text of the decision of the supreme court NJ 1991/593 07-12-1990
Supreme Court:
(…)
3. Evaluation of the means of cassation
3.1.1 In cassation the following must be taken as a starting point:
L who is born in 1944, has taken care of the 72-year old van Wylick from January 1983 receiving payment in compensation for the care, initially several days per week and in a later stage on a daily basis. On September 29, 1983 L has married mrs. Van Wylick without making a marriage settlement. The marriage took place in another place than where the future spouses lived and no publicity was given to the marriage.
L owned practically nothing while mrs. Van Wylick brought in a considerable fortune. Both knew that the marriage would cause a considerable shift of property.
Since 1976 L had a relation with another man, which relation has not been broken.
Five weeks after the marriage L has killed van Wylick in a sophisticated way and with a gross breach of the trust that had been put in him. L has been condemned to a long term imprisonment for murder.
3.1.2 Furthermore, on the basis of these circumstances, in particular the short time between the marriage and the murder of mrs. Van Wylick, in the absence of any offer of proof to the contrary, the court has taken as a starting point that the sole reason for L to marry mrs. van Wylick was that he intended to appropriate her property and that already during the wedding, and in any case almost immediately after, L had the intention to kill mrs. van Wylick if she would not die in a natural way.
3.1.3 The court of appeal has, in a similar way as the district court, ruled that the question whether L has a right to half of the property belonging to the community property in the context of the partitioning and division of the community property, as far as this is brought in by mrs. van Wylick, must be answered negatively. This decision is contested by the means of cassation.
3.2 In the legal consideration 5.10 the Court of Appeal has taken as a starting point in answering the aforementioned question that in the light of the ‘exceptional circumstances of this case’ on the one hand consideration must be given to the general legal principles and on the other hand to the requirements of reasonableness and fairness according to which L is supposed to behave in the community property.
Furthermore the court has stated in legal consideration 5.13-5.17 that in this case two general legal principles apply and that on the basis of these principles L is not entitled to the benefits that originate from the community property.  Against these two considerations the parts A and B of the means of cassation are aimed in vain.
As far as these parts are based on the statement that the general legal principles formulated by the court do not exist at all, this statement, that has not been substantiated, must be rejected as incorrect.
As far as these parts A and B are intended as an argument in support of the statement that these legal principles do not apply in a case as the case at hand because, briefly stated, the nature of the acquisition resulting from the community of property impedes that this acquisition can be considered as something that is equal to a ‘favour’ or an ‘ advantage’ as mentioned in these principles, they cannot lead to cassation because of a lack of interest. For the decision of the court is supported by the independent judgement formulated in consideration 5.18 that is, as will be explained below, contested in vain.

3.3 In legal consideration 5.18 the court has ruled that in the exceptional circumstances of this case ‘and also considered in light of the mentioned general legal principles’ the claims of L are so unreasonable and unfair that he must be denied the exertion of these rights completely. As appears from the cited formulation, in this context the legal principles play only the role that they have contributed to the decision of the court that the requirements of reasonableness and fairness make the exertion of the right to his share in the inheritance inadmissible. As far as the parts A and B read in legal consideration 5.18 that the court has used these principles as a direct legal ground for denying this right, they lack a factual basis. As far as they express the complaint that those principles cannot contribute to the decision of the court, they depart from a wrong conception of the law.
Part C attacks legal consideration 5.18 with the statement that the judge is not allowed to make an exception to 1:100, 1 of the Civil Code on the basis of reasonableness and fairness. This statement is wrong in its generality. For an exception is not completely excluded. The court has correctly stated that such an exception can only be made in very special circumstances, where the court speaks of ’ a very strict standard’ . In the circumstances that the court has taken as a starting point, the court has correctly decided that the unimpaired application of the equal division of the community of property based on the rule of article 1:100 clause 1 of the Civil Code between spouses in a dissolved matrimonial community, would, in the wording of article 6:2 clause 2 of the new Civil Code , be unacceptable according to standards of reasonableness and fairness.
On this ground the court has concluded that in the division of this community L is not entitled to the share in the community of property that has been brought in by van Wylick.
(…)

3.5 Since, as has been stated above, none of the parts succeed (‘treffen doel’), the appeal in cassation must be dismissed.

4. Decision
The Supreme Court:
dismisses the appeal;

« Qui donc crée de la valeur, à part les dieux? »
Édouard Tétreau, Analyste. Au cœur de la folie financière (2005, p. 62)

The present paper proposes an analysis of the argumentative use of the key-phrase value creation in corporate reporting discourse, in line with Rigotti and Rocci’s theoretical model of keywords as lexical pointers to unexpressed endoxa (2005). By means of a brief quantitative analysis of concordances conducted on a corpus of full-text reports, and a detailed argumentative analysis of a relevant sample of letters to shareholders (and stakeholders), the study attempts to grasp the main patterns of pragmatic meaning and argumentative moves prompted by value creation (as one single unit of meaning) in both annual reports and corporate social responsibility reports[i]. This twofold methodological approach will enable a concomitant focus on the two main keyness criteria envisaged by Stubbs’ generic definition of keywords as “words with a special status, either because they express important evaluative social meanings, or because they play a special role in a text or text-type” (in press, p.1).

1. Value creation in economic-financial discourse
In everyday language, value is an abstract notion that denotes the degree of worth and appreciation of a certain object, depending on its desirability or utility. The relative worth of an object can also be evaluated by the amount of things (e.g. goods or money) for which it can be exchanged, and this could be considered the departure point of the conceptual journey of value in the economic and financial fields.

From a strategic management perspective (Becerra 2009), the fundamental value created by a firm is the one created for its customers through the products and services that result from the judicious management of the available resources. This value is then (at least in part) appropriated by the firm through sales revenues, entering in the process of shareholder value creation. This process is aimed to increase the wealth of the owners of the company either directly, by dividends, or indirectly, by influencing, one way or another, the price of the shares – a price that reflects the perceived value of the company on the financial market, based on all its expected future cash flows (Schauten 2010).

From a business ethics point of view, there is, however, an ongoing debate on the type of value creation that should guide the managerial decisions in corporations (Smith 2003). On the one hand, the shareholder theory considers that the main duty of the managers is to maximize shareholders’ returns, and to spend the resources of the corporation only in ways that have been authorized by the shareholders. On the other hand, the stakeholder theory stresses that “a manager’s duty is to balance the shareholders’ financial interest against the interest of other stakeholders such as employees, customers and the local community, even if it reduces shareholder returns” (p.85).

An interesting instantiation of this debate can be observed in the way in which value creation is conceived and argumentatively exploited in corporate reporting. The (financial-economic) annual reports and the corporate social responsibility reports are publications by means of which listed corporations account for their activity in front of shareholders (and stakeholders at large), in order to build trustful relationships with current and potential investors, and to legitimate themselves as responsible citizens of the world, able to “meet the needs of the present without compromising the ability of future generations to meet their own needs” (World Commission on Environment and Development 1987, p. 43, in Global Reporting Initiative 2000-2006, p.2). Therefore, the present study will pay a special attention to the way in which value creation is reflected in these two types of reports, in particular in their most visible and influential narrative parts (Clarke & Murray 2000) – the introductory letters to shareholders and/or stakeholders.

2. Corpus description and methodological approach
The first phase of the study consisted of a brief computer-based analysis of a corpus of 26 financial-economic annual reports and 46 corporate social responsibility (sustainability) reports belonging to 22 listed multinational corporations. All reports referred to the financial year 2007 and were published on Internet on the websites of the respective companies[ii]. The purpose of this phase was to identify the generic pattern of pragmatic meaning of value creation in each category of reports, by means of Wordsmith Tools’ Concord analysis (considering value as search-word and creat* as context-word).

The second phase consisted of selecting from the above mentioned corpus of full-text reports, only those introductory letters (and in a limited number of cases, equivalent introductory interviews with CEOs or Presidents) that contained the key-phrase value creation. The selected documents (9 letters and 3 interviews from annual reports, and 7 letters, one introduction and 3 interviews from sustainability reports) were then argumentatively reconstructed in line with the pragma-dialectical principles (van Eemeren & Grootendorst 1999; Snoeck Henkemans 1997), in order to identify the main strategic moves in which value creation appeared. Next, a limited number of single argumentative moves were evaluated from the perspective of the Argumentum Model of Topics, in particular the taxonomy of loci (Rigotti 2008, 2006; Rigotti & Greco Morasso 2006-2010), in order to highlight the key-role of value creation (considered as one single unit of meaning) in line with Rigotti & Rocci’s model of argumentative cultural keywords (2005).

According to this model, culturally loaded words present in explicit minor premises may function, in virtue of their logical role of termini medii, as lexical pointers to shared values and beliefs (endoxa) that act as (implicit) major premises in support of certain claims. Paraphrasing Aristotle’s definition of endoxa as “opinions that are accepted by everyone or by the majority, or by the wise men (all of them or the majority, or by the most notable and illustrious of them)” (Topica 100b.21, in Rigotti 2006, p.527), Rigotti (2006) re-defines the endoxon as “an opinion that is accepted by the relevant public or by the opinion leaders of the relevant public” (p.527). Thus, a second characteristic of argumentative keywords consists in their persuasive potential – the capacity to evoke, from an (appropriately) assumed common ground, endoxa with different degrees of acceptability within certain communities.

3. Value creation in the introductory letters of the annual reports
The basic pattern of pragmatic meaning outlined by the most frequent concordances of value and creat* in the corpus of full-text annual reports shows that “Every company’s aim is to create value. To achieve this aim, decisions are taken and activities developed”. The value can be created “for the Company, for customers, and for the owners of the Company”, or “for [company’s] employees”, and generally speaking, “for all [company’s] stakeholders”. For instance “Heineken creates value and enjoyment for millions of people around the world […] through brewing”. The most frequently mentioned beneficiaries of the process of value creation are the shareholders, because “true value creation does translate into stock price appreciation”. Therefore, the possession of “a strong ability to create value in the different stages of the real estate market”, and promises such as “[our company] will create significant value from our assets in the years to come” are frequent arguments in this type of discourse aimed to win investor’s trust. As expected, various corporate resources are mentioned as material or operational base for value creation. For instance, a company may “create significant value from [its] assets”, “from eco-efficient solutions”, or “by earning higher margins” “through industry-leading performance”, and could do this “jointly with retail customers”.

Although value creation was present in almost all the annual reports of the corpus (in 22 out of 26 reports), the phrase appeared in the introductory letters of only half of them. The pattern of meaning observed in the full-text reports was also present in the letters, being included in a number of recurrent argumentative moves usually belonging to three main types of loci: the locus from final cause, the locus from efficient cause and the locus from instrumental cause.

As the main purpose of the annual reports is to attract (or keep) investors for the company, the (often implicit) standpoint of the introductory letters has the generic form: You should (continue to) invest in our company. The principal modality to support this standpoint is to show that an investment in the company can help shareholders to achieve their own ultimate goal which is to obtain good revenues from their investment (better than from other similar investment alternatives). A typical move in this direction is to highlight the good results obtained in the reporting year and to announce a (justified) optimistic outlook for the coming year, and the value created for the shareholders is the most frequent argument in this respect:

(1) “I am delighted to be able to report to you on another year of delivery of the Nestlé Model, defined as the achievement of a high level of organic growth together with a sustainable improvement in EBIT[iii] margin. […] We continue to believe that our greatest opportunity to create value for our shareholders is through further transforming our Food and Beverages business into a Nutrition, Health and Wellness offering and by improving its performance further. [The major steps in this transformation have now been made.] […] This is not to say, however, that we are not looking for other opportunities for value creation. (p.2) […] The Nestlé Model, combined with our ongoing ambitious Share Buy-Back Programme, will deliver strong earnings per share growth [in the coming year], resulting in industry-outperforming, long-term shareholder value creation.”(p. 5)
(Letter to our shareholders. Nestlé Management Report 2007: Life.)

A simplified reconstruction of this sample of pragmatic argumentation could be:
(2)  (SP) (You should invest in Nestlé.)
(1) (Your goal, as a shareholder, is to have a (good) return on your investment.)
(1’) (Investing in Nestlé enables you to reach your financial goal.)
1’.1a We have created value for our shareholders in 2007.
1’.1b In 2008 we will create industry-outperforming, long-term shareholder value.

We recognize in this structure the locus from final cause (Rigotti 2008), that I represent below according to the Argumentum Model of Topics, and in which value creation has the role of terminus medius between the explicit Datum and the implicit Endoxon evoked from the context (on the left side of the Y-shaped structure):

Two other types of moves are used in the letters to shareholders in annual reports, in order to support the claim that an investment in the company would help investors to achieve their own final goal. The first type of moves regards the agency relationship between the company and its shareholders, and it is based on the locus from efficient cause. The second emphasizes the quality of the means employed by the company in order to accomplish its task, and it makes use of the locus from instrumental cause.    

For instance, if we add to the argumentative structure represented in the Example no.2 the endoxon (1’.1.1’(a-b)) evoked from the corporate context by the key-phrase value creation, we can underline the fact that the value created by the company for its shareholders is a proof of the reliability of the company in relation to its shareholders:
(3) (SP) (You should invest in Nestlé.)
(1) (Your goal, as shareholders, is to have a (good) return for your investment.)
(1’) (You can rely on Nestlé in order to reach your financial goal.)
(1’.1) (We fulfil our mission towards our shareholders.)
1’.1.1a  We have created value for our shareholders in 2007.
1’.1.1b  In 2008 we will create industry-outperforming, long-term shareholder value.
(1’.1.1’(a-b)) (The mission of a company is to create value for its shareholders.)
(1’.1’) (An agent that fulfils its mission towards its principal is reliable.)

Based on the same locus from efficient cause, the value created for the shareholders can be an argument in support of the unique managerial capabilities of the company, given that in business, uniqueness is a source of competitive advantage:
(4) “Or, you could pick GE. (p.1) […] GE is different because we invest in the future and deliver today. […] We are a leadership company. We have built strong businesses that win in the market. (p.2)

[…] “We develop leadership businesses. […] In 2007, we demonstrated the ability to create value for our investors through capital redeployment. We sold our Plastics business because of rampant inflation in raw material costs. With that capital we acquired Vetco Gray […]. We significantly exceeded the earnings we lost from Plastics, increased our industrial growth rate, and launched new platforms for future expansion.” (p.5)
(Letter to investors. GE Annual Report 2007: Invest and Deliver Every Day.)

Textual clues indicate that the two fragments extracted from different sections of the above introductory letter can be interpreted as parts of the same line of argumentation, as follows:
(5) SP You should pick (invest in) GE.
(1) (Your goal, as shareholders, is to have a (good) return for your investment.)
(1’) (Investing in GE enables you to reach your financial goal.)
1’.1  GE is different.
1’.1.1  We invest in the future and deliver today.
1’.1.1.1 We develop leadership businesses.
1’.1.1.1.1 In 2007, we demonstrated the ability to create value for our investors through capital redeployment.
(1’.1’)  (Uniqueness is a source of competitive advantage.)

The value created for investors in the reporting year becomes an argument for the market leadership of GE’s businesses, and further on, for the ability of the company to “invest in the future and deliver today”. Creating value from capital redeployment signifies delivering results today (short-term value creation) from sound strategic choices of acquisitions and divestitures of businesses, and investing in the future of the company (preparing the portfolio for medium and long-term value creation).

The quality of the strategy that guides the managerial choices leads us to the second main category of argumentative moves used in support of the ability of companies to benefit shareholders: the possession of the “right” means (locus from instrumental cause). As resulting from the corpus of letters, the main argument for the soundness of a strategy is its capacity to enable shareholder value creation. The emphasis can be placed either on the value creation potential of the business strategy as a whole, like in Example no.6 below:
(6) “[…] our greatest opportunity to create value for our shareholders is through further transforming our [business] and by improving its performance further. We believe that we have the right strategy and initiatives in place to achieve this.”
(Letter to our shareholders. Nestlé Management Report 2007: Life, p. 4)

or on the value creation potential of single strategic steps:
(7)  “Through an on sale of certain ICI assets to Henkel AG, we expect the acquisition to be value enhancing within three years. This is fully in line with our strategic goal of medium-term value creation.”
(Chairman’s statement. Akzo Nobel Annual Report 2007: Year of Transformation, p. 12)

The value creation potential of the business strategy can also be strategically manoeuvred in order to defend the status quo of the strategy itself. In this final example extracted from the introductory letters of the annual reports, a CEO must face shareholders’ (potential) critiques on the distribution of the profits:

(8) “[Question]: PepsiCo’s businesses generate a lot of cash, and some people may believe the company’s balance sheet is conservative. Will investors see any changes in capital structure, acquisition activity or increased share repurchases?

[Answer]: PepsiCo does generate considerable cash, and we are disciplined about how cash is reinvested in the business. Over the past three years, over $6 billion has been reinvested in the businesses through capital expenditures to fuel growth. All cash not reinvested in the business is returned to our shareholders. […] We will generally use our borrowing capacity in order to fund acquisitions — which was the case in 2007, when we spent $1.3 billion in acquisitions to enhance our future growth and create value for our shareholders. Our current capital structure and debt ratings give us ready access to capital markets and keep our cost of borrowing down.”
(Questions and Answers: A Perspective from Our Chairman and CEO. PepsiCo 2007 Annual Report: Performance with Purpose. The Journey Continues…, p. 9)

The CEO refutes the possible negative connotations of an unchanged financial strategy (suggested in the question by the risk of being perceived as conservative) by highlighting the benefits of that strategy for the shareholders:

(9) SP We will not make changes in the current capital structure, acquisition   activity or shares repurchase.
1  Our current strategy is valuable for the shareholders.
1.1a PepsiCo generates a lot of cash.
1.1b We are disciplined about how cash is reinvested in the business.
1.1b.1a All the money reinvested was used to fuel growth (i.e. for future value creation).
1.1b.1b All that remaining cash was returned to shareholders (it created value for the shareholders).
(1.1c) (The current financial strategy allows us to continue to create value for our shareholders in the future.)
1.1c.1a Our current capital structure gives us ready access to capital markets and keeps our cost of borrowing down.
1.1c.1b We use our borrowing capacity in order to fund acquisitions.
1.1c.1b’ Acquisitions enhance our future growth and create value for our shareholders.
(1.1’(a-c)) (If a strategy produces valuable effects, then that strategy is valuable.)
(1’) (If a strategy is valuable for the shareholders, then that strategy should not be changed.)

In order to prove that the current financial strategy is valuable, the CEO tactically chooses to underline not only the value created for the shareholders in the reporting year, but also the expected value that can be created in the future by following this strategy (locus from the instrumental cause, indicated by the premise (1.1’(a-c)). This topical choice is aimed to support the fact that a change of the financial strategy is not necessary, as it would be unreasonable for shareholders to ask for a change in a strategy that has already brought them benefits and it will also enable them to obtain future benefits (the locus from termination and setting up, indicated by the premise (1’)). This could be an effective manoeuvre, unless shareholders have different expectations for the revenues they obtain from their investment (e.g. a preference for immediate short-term gains rather than medium or long-term gains).

As a final observation, I must add that manually checking a random sample of full-text annual reports, I have noticed that the phrase value creation appears only in the narrative sections, and not in the proper financial sections of the reports. That suggests that although (shareholder) value creation is invoked in this category of letters as “the primary measure of business and financial performance” (P&G Annual Report 2007, p.5), the expression does not directly denote an objective financial indicator, its function being mainly rhetorical.

4. Value creation in the introductory letters of the corporate social responsibility reports
The same analytical steps have been followed in the study of the corporate social responsibility reports and the related letters to stakeholders. Like in the case of the annual reports, value creation appeared in 70% of the reports of the corpus, but only in half of their introductory letters.

The pattern of pragmatic meaning outlined by the main recurrent concordances of value and creat* in the corpus of full-text reports shows that value creation maintains its strategic role in this new type of discourse: (“Every company’s aim is to create value. To achieve this aim, decisions are taken and activities developed”). However, the scope of the phrase is extended in terms of presupposed activities, results and beneficiaries: “[we] achieve optimal performance and create sustainable value for all [our] stakeholders” (for employees, customers, communities, governments, society at large) and “for the planet” (the environment). The commitment to sustainability starts at the level of process (“[we] create value by observing the business world from a new perspective”, “through genuine partnership with stakeholders” (customers, communities and governments), and continues up to the level of business effects:  “[companies] with distinctive capabilities to create eco-efficient sustainable value will be the winners in the more demanding global market place”, because “developing a relationship with communities does not only create value for them but also contributes to the company’s value“ and “This is both a commercial and CR [Corporate Responsibility] win–win.”.

This pattern is confirmed by the way in which value creation is argumentatively employed in the sub-corpus of letters to stakeholders selected from this type of reports. Being representative of a reporting genre aimed at legitimizing corporations as responsible members of the society, the generic standpoint of the letters to stakeholders is a declaration or a reinforcement of the commitment of the corporations to social responsibility, to the fulfilment of their obligations towards society. Although the precise way in which these obligations are seen may differ from one company to another, the main topics of social responsibility presented in the corpus of letters generally comply with the (deontological) norms of voluntary disclosures on sustainability recommended by the Global Reporting Initiative (2000-2006).

Value creation maintains the supremacy among the corporate goals mentioned in this type of letters, but the range of beneficiaries and constituent activities is significantly extended. An illustrative example is presented below:
(10) “Creating Shared Value: the role of the business in society. […] the fundamental strategy of our Company has been to create value for society, and in doing so create value for our shareholders. […] Creating Shared Value for society and investors means going beyond consumer benefit. […] Creating Shared Value also means bringing value to the farmers who are our suppliers, to our employees, and to other parts of society. It means examining the multiple points where we touch society and making very long-term investments that both benefit the public and benefit our shareholders, who are primarily pension savers or retirees. […] Creating Shared Value additionally means treating the environment in a way that preserves it as the basis of our business for decades, and centuries, to come. […] Creating Shared Value means thinking long term, while at the same time delivering strong annual results. One of the fundamental Nestlé Corporate Business Principles is that ‘we will not sacrifice long-term development for short-term gain’.”
(Creating Shared Value: the role of business in society. The Nestlé Creating Shared Value Report, p. 2)

Basically, the argumentative structure of the Example no.10 can be reconstructed as follows:
(11) (SP) (We are a socially responsible corporation.)
1 We accomplish our role in society.
1.1 We Create Shared Value for society and investors.
(1.1.1a) (Creating Shared Value means bringing benefits to consumers.)
1.1.1b Creating Shared Value means examining the multiple points where we touch society and making very long-term investments that both benefit the public and benefit our shareholders.
1.1.1c Creating Shared Value means bringing value to the farmers who are our suppliers, to our employees, and to other parts of society.
1.1.1d Creating Shared Value means treating the environment in a way that preserves it as the basis of our business for decades, and centuries to come.
1.1.1e Creating Shared Value means thinking long term, while at the same time delivering strong annual results.
1.1.1f [We do all these things.]
(1.1.1f ’) (An entity can be defined with a certain property, if it satisfies (all) the necessary conditions for that property.)
1.1’ The role of the business in society is to Create Shared Value for society and investors.                                          
(1’) (If a company accomplishes its role in society, then that company is socially responsible.)

To prove that it is a socially responsible company, Nestlé shows that it accomplishes the main role of a business in society, i.e. its duty towards society (the locus from efficient cause). In order to do that, Nestlé presents its own vision of the role of the business in society (to Create Shared Value), and strategically defines this new type of value creation by providing a number of necessary conditions related to sustainability that should be satisfied by any socially responsible business. Facts from the reality of the company are then provided in order to prove that all these conditions are satisfied – proofs generically marked in the structure above by the premise 1.1.1f. Thus, in virtue of a complex locus from definition and from the parts and the whole, indicated by premise (1.1.1f ’) – maxim adapted from Rigotti & Greco Morasso 2006-2010[iv] – the company proves that it creates Shared Value; hence, it can be considered socially responsible.

The whole construction of the concept of Shared Value Creation could be considered a persuasive definition (Stevenson 1938; Macagno & Walton 2010) aimed to introduce Nestlé’s vision of the role of the business in society (premise 1.1’) as an already accepted endoxon, without necessarily defending it. The persuasive mechanism of this definition would consist in the transfer of the strong positive connotations acquired by (shareholder) value creation in financial-economic discourse (generally accepted as the aim of a corporation, rigorously implemented and highly appreciated by the target-beneficiaries), to the different, far less regulated domain of sustainability that presupposes different types of activities (some of them still based on voluntarism), and that envisages a wide range of results (not all clearly measurable) and a heterogeneous set of beneficiaries (and expectations). The substitution of the qualifier shareholder with shared in the definition of the new concept of value creation, could also have a peripheral effect of reinforcement of the positive emotions elicited by value creation in this new context. But the true meaning of shared is further (indirectly) indicated in the text by the arguments that prove that the company Creates Shared Value through the activities described in premises 1.1.1a – 1.1.1e. In fact, the social and environmental effects of these activities would eventually benefit the company itself. Two conclusions can be drawn from this aspect. Firstly, the concept of Creating Shared Value, as operationalized in the text, may be a good definition of the role of Nestlé in society, but not of the role of business in general, in which case the premise 1.1’ from Example no. 11 cannot be used as an endoxon. Secondly, even if the premise 1.1’ refers to a general principle that connects social value with corporate performance in terms of moral duty or in terms of business opportunity, the argumentation provided in the text is insufficient in order to consider this premise an endoxon (a generally accepted opinion on the role of business in society) in either way.

As resulting from the sub-corpus of letters to stakeholders, there is however a tendency to use the shareholder value creation potential of sustainability as an argument of socially responsible corporate behaviour, like in Example no.12:
(12) “[Sustainability] is at the center of our strategy and rightfully so. […] [It] contributes to growth and value creation. Initially people thought of it as a cost factor, which indeed it is when you treat it as an add-on. However, if it’s designed into the way you do things from the beginning as it is here at Philips, it saves you money because you’re operating more effectively. So today we recognize that sustainability offers significant business opportunities.”
(Interview with the president. Philips Sustainability Report 2007: Simpler, stronger, greener, p. 8)

In this example, Philips’ president highlights the value-creation opportunities offered by sustainability if it is approached with the “right” managerial attitude (e.g. taking sustainability as the departure point for the production of goods), as opposed to the “wrong” managerial attitude (e.g. superficially implementing it, considering it an add-on):
(13) (SP) (We (will) behave sustainably.)
1 Sustainability contributes to growth and value creation.
1.1a Sustainability offers significant business opportunities.
1.1b Sustainability is not a cost factor.
1.1b.1 Sustainability saves us money.
(1’) (Every company’s aim is to create value.)
(1’’) (If an action contributes to the achievement of a desired goal, then that action should be undertaken.) [Pragmatic argumentation – locus from final cause]

Or, alternatively:
(1’’’) (An agent’s commitment is reliable if it is bound to its strongest interest.) [Locus from efficient cause] (maxim quoted from Rigotti, Greco Morasso, C. Palmieri & R. Palmieri 2007[v])

I will represent the latter alternative by means of the Argumentum Model of Topics. As in Example no.2, the premise (1’) is an implicit endoxon evoked from the context by the key-phrase value creation:

On the other hand, shareholder value creation, as ultimate corporate aim, can be used as an excuse for not meeting the (excessive) expectations of the stakeholders towards the company, as in the next example:

(14) “Businesses have to be honest about what they are and what they can do. Our goal is to create sustainable shareholder value. Businesses can’t assume the role of governments, charities, political parties, action groups or the many other bodies that make up society.”
(Chief Executive’s Overview. British American Tobacco Sustainability Report 2007, p. 3)

Example no.14 can be interpreted as follows, by means of the locus from final cause – indicated by premise (1.1’) below, and the locus from the parts and the whole – indicated by premise (1’):

(15) (SP) (We cannot resolve (alone) all the sustainability issues of the society.)
1 We cannot assume the role of governments, charities, political parties, action groups or the many other bodies that make up society.
1.1 Our goal is to create sustainable shareholder value.
(1.1’) (A company cannot (be reasonably expected to) assume roles that are not related to its final goal.)
(1’) In order to resolve all the sustainability issues of the society, all social partners must assume their role.

British American Tobacco continues, however, its discourse by constructing a “business case for sustainability” based on the contribution of sustainability to corporate performance, similar to the “win-win” move presented in Example no.13.

5. Conclusions
The purpose of this corpus-based study was to observe the argumentative use of the key-phrase value creation in corporate reporting, by a comparison between the letters to shareholders from the annual reports, and the letters to stakeholders from the corporate social responsibility reports. The analytical tools employed in the study confirmed the status of key-phrase for value creation (as one single unit of meaning), in line with Stubbs’ generic definition of keywords as “words with a special status, either because they express important evaluative social meanings, or because they play a special role in a text or text-type” (in press, p.1). A frequent occurrence in the corpus, value creation was proven to have genre-specific denotative and evaluative meanings, illustrative for the corporate goals, activities and relationships with the stakeholders, thus complying with Williams’ idea of cultural keywords as “[…] significant, binding words in certain activities and their interpretation” (1976, p.13, in Bigi 2006, p.163).

Defining the essence of the agency relationship between corporation and different categories of stakeholders (especially with the shareholders), value creation was frequently used as an argument in both types of letters, usually in close proximity to the principal standpoint of the letter. Complying with Rigotti and Rocci’s model of argumentative keyword, value creation evoked two main goal-related categories of endoxa. The first category stressed the final goal of the shareholders (or stakeholders at large): to obtain what they want (request) from a corporation; the second category stressed the final goal of the corporations: to fulfil their mission towards stakeholders, by providing what they have been asked to provide.

As expected, the ethical debate between the shareholder theory and the stakeholder theory (previously illustrated in Chapter 1) was evident in the argumentation of the two categories of introductory letters. In annual reports, the letters emphasized the ability of a corporation to create value for the shareholders (through unique management qualities or/and the right means) as main argument in order attract investors. Accordingly, the basic argumentative pattern prompted by value creation consisted of a principal move based on the locus from final cause, supported by arguments from efficient cause and from instrumental cause. Although shareholder value creation was considered the ultimate corporate aim in both types of reports, and shareholders were considered the most important stakeholders, a series of attempts to unify the two opposite ethical views (at least at the level of discourse) were observed in the corpus, especially in the letters to stakeholders from the corporate responsibility reports. A first category of attempts was based on the semantic shift of value creation from the financial domain to the domain of social responsibility, Example no.10 being representative in this respect. The second category, most frequently encountered, was based on pragmatic argumentation, viewing sustainability as a potential source of shareholder value creation. Thus, corporations could reasonably be expected to behave sustainably as long as this is in their own best interest – a “win-win” strategy. In my opinion this move, that belongs to the locus from efficient cause, is the most representative for the letters to stakeholders in social corporate responsibility reports.

The intention of this study was not to question the conceptual and ethical approach to value creation of various theories of the firm, but to see how value creation is pragmatically reflected and argumentatively exploited in two sub-genres of persuasive business discourse: the introductory letters to shareholders and stakeholders, from the annual, respectively, corporate social responsibility reports. Although the examples presented in this paper did not exhaust all the argumentative instances of value creation in the corpus letters, I hope that they offered some useful insights on this topic.

NOTES
[i] This study was developed within the framework of the project “Endoxa and keywords in the pragmatics of argumentative discourse. The pragmatic functioning and persuasive exploitation of keywords in corporate reporting”, funded by the Swiss National Science Foundation (Grant SNSF PDFMP1_124845/1) and coordinated by Andrea Rocci at Università della Svizzera italiana in Lugano.
[ii] All the reports included in the corpus were published in .pdf format on the websites of the correspondent companies, being identical with the homonymous printed documents.
[iii] EBIT – earnings before interests and taxes.
[iv] The premise (1.1.1f ’) from the argumentative reconstruction of Example no.11 partially reproduces a maxim included in an example of locus from the Argumentum eLearning Module (Rigotti & Greco Morasso 2006-2010).
[v] The premise (1’’’) from the reconstruction of Example no.13 integrally reproduces a maxim included in an example of locus from the e-course Argumentation for Financial Communication, the Argumentum eLearning Module (Rigotti, Greco Morasso, C. Palmieri. & R. Palmieri 2007).

REFERENCES
Becerra, M. (2009). Theory of the Firm for Strategic Management: Economic Value Analysis. Cambridge: Cambridge University Press.
Bigi, S. (2006). Focus on cultural keywords. Studies in Communication Sciences, 6(1), 157-174.
Clarke, G., & Murray L.W.(B.) (2000). Investor relations: Perceptions of the annual statement. Corporate Communication: An International Journal, 5(3), 144-151.
Eemeren, F.H. van, & Grootendorst, R. (1999). From analysis to presentation: A pragma-dialectical approach to writing argumentative texts. In J. Andriessen & P. Coirier (Eds.), Foundations of Argumentative Text Processing (pp. 59-73). Amsterdam: Amsterdam University Press.
Global Reporting Initiative (2000-2006). RG. Sustainability Reporting Guidelines, Version 3.0, 1-45. Retrieved August 1, 2010, from http://www.globalreporting.org/NR/rdonlyres/ED9E9B36-AB54-4DE1-BFF2-5F735235CA44/0/G3_GuidelinesENU.pdf
Macagno, F., & Walton, D. (2010). What we hide in words: Emotive words and persuasive definitions. Journal of Pragmatics, 42, 1997-2013.
Rigotti, E. (2006). Relevance of context-bound loci to topical potential in the argumentation stage. Argumentation, 20(4), 519-540.
Rigotti, E. (2008). Locus a causa finali. In G. Gobber, S. Cantarini, S. Cigada, M.C. Gatti & S. Gilardoni (Eds.), L’analisi linguistica e letteraria, XVI (2), Special Issue: Word meaning in argumentative dialogue (pp. 559-576). Milan: Facoltà di Lingue e Letterature straniere, Università Cattolica del Sacro Cuore.
Rigotti, E., & Greco Morasso, S. (2006-2010). Argumentum eLearning Module. Università della Svizzera italiana, Lugano. Retrieved from http://www.argumentum.ch.
Rigotti, E., Greco Morasso, S., Palmieri, C., & Palmieri, R. (2007). Argumentation for Financial Communication [Electronic course], Argumentum eLearning Module, Università della Svizzera italiana, Lugano. Retrieved March 18, 2007, from http://www.argumentum.ch.
Rigotti, E., & Rocci, A. (2005). From argument analysis to cultural keywords (and back again). In F.H. van Eemeren & P. Houtlosser (Eds.), Argumentation in Practice (pp. 125-142). Amsterdam: John Benjamins.
Schauten, M. (2010). Shareholder value misunderstood. Valuation Strategies, 13(3), 34-35.
Smith, H. J. (2003). The shareholders vs. stakeholders debate. MIT Sloan Management Review, 44(4), 85-90.
Snoeck Henkemans, A.F. (1997). Analyzing Complex Argumentation: The Reconstruction of Multiple and Coordinatively Compound Argumentation in a Critical Discussion. Amsterdam: Sic Sat.
Stevenson, C.L. (1938). Persuasive definitions. Mind, 47(187), 331-350.
Stubbs, M. (in press). Three concepts of keywords. In M. Bondi & M. Scott (Eds.), Keyness in Texts: Corpus Linguistic Investigations (pp. 21-42). Amsterdam: John Benjamins. Pre-publication version (1-19) retrieved on August 1, 2010, from http://www.uni-trier.de/fileadmin/fb2/ANG/Linguistik/Stubbs/stubbs-2008-keywords.pdf.
Tetreau, É. (2005). Analyste: Au cœur de la folie financière. Paris: Bernard Grasset.

Corpus of corporate reports
a) Annual reports:
Akzo Nobel N.V. (2008). Annual Report 2007: Year of Transformation. Amsterdam: Akzo Nobel Corporate Communications.
Audi AG (2008). Annual Report 2007: Passion; Success Formula for a Mobile Lifestyle. Ingolstadt: Audi AG.
British American Tobacco (2008). Annual Report and Accounts 2007. London: British American Tobacco.
BP plc (2008). BP Annual Report and Accounts 2007. London: BP plc
Canon Inc. (2008). Canon Annual Report 2007. Tokyo: Canon Inc.
Entergy Corporation (2008). 2007 Annual Report: Energy Corporation Presents: “Unlocking Value. Cracking Even the Toughest of Cases!” New Orleans: Entergy Corporation.
General Electric Company (2008). GE Annual Report 2007: Invest and Deliver Every Day. Fairfield: General Electric Company.
GlaxoSmithKline plc (2008). Annual Review 2007: Answering the Questions that Matter. Brentford: GlaxoSmithKline plc.
Heineken N.V. (2008). 2007 Annual Report. Amsterdam: Heineken N.V. Group Corporate Relations.
KBC Group N.V. (2008). Annual Report 2007. Brussels: KBC Group N.V.
Melco International Development Limited (2008). Annual Report 2007: Infinite Vision. Hong Kong/ Macau: Melco International Development Limited.
Nestlé S.A. (2008). Management Report 2007: Life. Cham/ Vevey: Nestlé S.A.
PepsiCo, Inc. (2008). 2007 Annual Report: Performance with Purpose; the Journey Continues… Purchase: PepsiCo, Inc.
Pirelli & C. S.p.A. (2008). Directors’ report and consolidated financial statement at December 31, 2007. Milan: Pirelli & C. S.p.A.
Pirelli & C. Real Estate S.p.A. (2008). Annual Report 2007: Building the Future. Milan: Pirelli & C. Real Estate S.p.A.
The Procter & Gamble Company (P&G) (2007). 2007 Annual Report: Designed to Grow. Cincinnati: The Procter & Gamble Company.
PT Antam Tbk (2008). Annual Report for the Year 2007: Generating Higher Returns for a Better Future. Jakarta: PT Antam Tbk.
Royal Philips Electronics (2008). Philips Annual Report 2007: Simpler, Stronger, Better. Amsterdam: Koninklijke Philips Electronics N.V.
Sinopec Corp. – China Petroleum & Chemical Corporation (2008). 2007 Annual Report and Accounts. Beijing: Sinopec Corp. – China Petroleum & Chemical Corporation.
Škoda Auto a.s. (2008). Skoda Auto 2007 Annual Report: Cars and People. Mladá Boleslav: Škoda Auto a.s.
Vodafone Group plc (2008). Annual Report for the year ended 31March 2008. Newbury: Vodafone Group plc.
Volkswagen AG (2008). Annual Report 2007: Driving Ideas. Wolfsburg: Volkswagen AG.
Volkswagen Bank GmbH (2008). Annual Report 2007. Braunschweig: Volkswagen Bank GmbH.
Volkswagen Financial Services AG (2008). Annual Report 2007. Braunschweig: Volkswagen Financial Services AG.
Volkswagen Leasing GmbH (2008). Annual Report 2007. Braunschweig: Volkswagen Leasing GmbH.
Xerox Corporation (2008). Annual Report 2007: Listening, Connected, Committed to…You. Norwalk: Xerox Corporation.
Xstrata plc (2008). Annual Report 2007. Zug: Xstrata plc.

b) Corporate social responsibility reports:
Akzo Nobel N.V. (2008). Sustainability Report 2007: Year of Transformation. Amsterdam: Akzo Nobel Corporate Communications.
British American Tobacco (2008). Sustainability Report 2007. London: British American Tobacco.
BP plc (2008). Sustainability Report 2007. London: BP plc.
BP Australia Pty Limited (2008). BP in Australia: Sustainability Review 2007-08. Melbourne: BP plc.
BP Azerbaijan SPU (2008). BP in Azerbaijan: Sustainability Report 2007. Baku: BP Azerbaijan SPU.
BP Georgia (2008). BP in Georgia: Sustainability Report 2007. Tbilisi: BP Georgia.
Canon Inc. (2008). Canon Sustainability Report 2008. Tokyo: Canon Inc.
Entergy Corporation (2008). Sustainability Report 2007: Energy Corporation Presents: “Unlocking Value”. Extended Indefinitely! In Environmental, Social and Economic Performance Theatres Everywhere. New Orleans: Entergy Corporation.
General Electric Company (2008). GE Citizenship Report 2007-200: Investing and Delivering in Citizenship. Fairfield: General Electric Company.
General Electric Company (2008). GE ecomagination Report 2007: Investing and Delivering on ecomagination. Fairfield: General Electric Company.
GlaxoSmithKline plc (2008). Corporate Responsibility Report 2007: Answering the Questions that Matter. Brentford: GlaxoSmithKline plc.
Heineken N.V. (2008). 2007 Sustainability Report: Preparing for Our Future. Amsterdam: Heineken N.V. Group Corporate Relations.
KBC Group N.V. (2008). Corporate Social Responsibility Report 2007. Brussels: KBC Group N.V.
Melco International Development Limited (2008). Corporate Social Responsibility Report 2007. Hong Kong/ Macau: Melco International Development Limited.
Nestlé S.A. (2008). The Nestlé “Creating Shared Value” Report 2007. Vevey: Nestlé S.A., Public Affairs.
PepsiCo, Inc. (2008). Overview2007: Our Sustainability Journey. Purchase: PepsiCo, Inc.
Royal Philips Electronics (2008). Sustainability Report 2007: Simpler, Stronger, Greener. Amsterdam: Koninklijke Philips Electronics N.V.
Pirelli & C. S.p.A. (2008). Sustainability Report. In: Pirelli & C. S.p.A., Directors’ report and consolidated financial statement at December 31, 2007 (pp. 177-296). Milan: Pirelli & C. S.p.A.
Pirelli & C. Real Estate S.p.A. (2008). Sustainability Report 2007: Building the Future. Milan: Pirelli & C. Real Estate S.p.A., Communications Department.
The Procter & Gamble Company (P&G) (2007). 2007 Global Sustainability Report: Designed to Grow…Sustainably. Cincinnati: The Procter & Gamble Company.
PT Antam Tbk (2008). Sustainability Report, Year 2007: Operating Sustainably. Jakarta: PT Antam Tbk.
Sinopec Corp. – China Petroleum & Chemical Corporation (2008). Sustainable Development Report 2007. Beijing: Sinopec Corp. – China Petroleum & Chemical Corporation.
Škoda Auto a.s. (2008). Sustainability Report 2007/2008. Mladá Boleslav: Škoda Auto a.s.
Vodafone Australia Limited (2008). Corporate Responsibility Report 2008: Connect; Make the Most of Now. Chatswood: Vodafone Australia Limited.
Vodafone Group Plc. (2008). Corporate Responsibility Report for the year ended 31March 2008: One strategy. Newbury: Vodafone Group Plc.
Vodafone Malta Limited (2008). Corporate Social Responsibility April 2006 – March 2008. Birkirkara: Vodafone Malta Limited.
Vodafone New Zeeland Limited (2008). Corporate Responsibility Report 2008 for the 2008 Financial Year: Open; 10 Years of Vodafone New Zeeland. Auckland: Vodafone New Zeeland Limited.
Vodafone Portugal – Comunicações Pessoais S.A. (2008). Social Responsibility Report 07. Lisboa: Vodafone Portugal – Comunicações Pessoais S.A.
Vodafone Spain (2008). Corporate Responsibility Report 2007-08: Responsible Innovation. Madrid: Vodafone España S.A.
Vodafone UK (2008). CR review. Vodafone UK Corporate Responsibility 2007/08.
Newbury: Vodafone UK.
Volkswagen AG (2007). Sustainability Report 2007/2008: We Are Moving into the Future Responsibly. Wolfsburg: Volkswagen AG, Environment.
Xerox Corporation (2007). 2007 Report on Global Citizenship. Our Word. Our Work. Our World. Norwalk: Xerox Corporation.
Xstrata Alloys (2008). Sustainability Report 2007. Rustenburg: Xstrata South Africa (Pty) Limited Alloys.
Xstrata Copper – El Pachón Project (2008). Sustainability Report 2007. Las Condes: El Pachón Project. Capital San Juan: Xstrata Copper Chile SA.
Xstrata Copper – Frieda River Project (2008). Sustainability Report 2007. Brisbane: Xstrata Queensland Limited. Port Moresby NCD: Frieda River Project Registered Office.
Xstrata Copper – Tampakan Project (2008). Sustainability Report 2007. Tampakan: Tampakan Office. General Santos City: General Santos City Office. Makati City: Sagittarius Mines Inc., Head Office.
Xstrata Copper – Townsville (2008). Sustainability Report 2007. Brisbane: Xstrata Queensland Limited. Townsville: Copper Refineries Pty Ltd.
Xstrata Copper Canada (2008). Sustainability Report 2007.Toronto: Xstrata Copper Canada.
Xstrata Copper North Chile Division (2008). 2007 Sustainability Report. Antofagasta: Xstrata Copper North Chile Division.
Xstrata Copper Southern Peru Division – Las Bambas Mining Project (2008). Sustainability Report 2007. Chacarilla del Estanque: Xstrata Perù S.A.
Xstrata Copper Southern Peru Division – Xstrata Tintaya (2008). Sustainability Report 2007. Vallecito: Xstrata Tintaya S.A.
Xstrata Minera Alumbrera (2008). Sustainability Report 2007. Distrito de Hualfin: Minera Alumbrera.
Xstrata Mount Isa Mines (2008). Sustainability Report 2007. Brisbane: Xstrata Queensland Limited.
Xstrata Nickel (2008). Making Sustainable Progress. Toronto: Xstrata Nickel.
Xstrata plc (2008). Sustainability Report 2007. Zug: Xstrata plc.
Xstrata Zinc North Queensland (2008). Sustainability Report 2007. Brisbane: Xstrata Queensland Limited.

Sub-corpus of introductory messages
a) Documents extracted from the annual reports:
Brabeck-Letmathe, P. (2008). Letter to our shareholders. In: Nestlé S.A., Management Report 2007: Life (pp. 2-5). Cham/ Vevey: Nestlé S.A.
Davis, M.L. (2008). Chief Executive’s report. In: Xstrata plc, Annual Report 2007 (pp. 5-13). Zug: Xstrata plc.
GlaxoSmithKline plc (2008). Company performance (Question three. Share prices in the sector haven’t performed well, what is the outlook for GSK?). In: GlaxoSmithKline plc, Annual Review 2007: Answering the Questions that Matter (pp. 6-7). Brentford: GlaxoSmithKline plc.
Guatri, L. (2008). Chairman’s letter. In: Pirelli & C. S.p.A., Directors’ report and consolidated financial statement at December 31, 2007 (pp. 12-13). Milan: Pirelli & C. S.p.A.
Immelt, J. R.(2008). Letter to investors. In: General Electric Company, GE Annual Report 2007: Invest and Deliver Every Day (pp. 1-9). Fairfield: General Electric Company.
Kleisterlee, G. (2008). Message from the President. In: Royal Philips Electronics, Philips Annual Report 2007: Simpler, Stronger, Better (pp. 10-15). Amsterdam: Koninklijke Philips Electronics N.V.   
Lafley, A.G. (2007). Letter to shareholders. In: The Procter & Gamble Company, 2007 Annual Report: Designed to Grow (pp. 2-7). Cincinnati: The Procter & Gamble Company.
Leonard, J.W. (2008). Letter to stakeholders. In: Entergy Corporation, 2007 Annual Report: Energy Corporation Presents “Unlocking Value. Cracking even the toughest of cases!” (pp. 2-6), New Orleans: Entergy Corporation.
PepsiCo, Inc. (2008). Questions and answers: A perspective from our Chairman and CEO. In: PepsiCo, Inc., 2007 Annual Report: Performance with Purpose; the Journey Continues… (pp. 7-9). Purchase: PepsiCo, Inc.
Puri Negri, C.A. (2008). Message from the CEO. In: Pirelli & C. Real Estate S.p.A., Annual Report 2007: Building the Future (pp. 12-13). Milan: Pirelli & C. Real Estate S.p.A.
Volkswagen AG (2008). Talking to. (Prof. Dr. Martin Winterkorn is interviewed by Dirk Maxeiner). In: Volkswagen AG, Annual Report 2007: Driving Ideas (pp. 14-17). Wolfsburg: Volkswagen AG.
Wijers, H. (2008). Chairman’s statement. In: Akzo Nobel N.V, Annual Report 2007: Year of Transformation (pp. 12-14). Amsterdam: Akzo Nobel Corporate Communications.

b) Documents extracted from the corporate social responsibility reports:
Adams, P. (2008). Chief Executive’s overview. In: British American Tobacco, Sustainability Report 2007 (pp. 2-3). London: British American Tobacco.
Brabeck-Letmathe, P. (2008). Creating Shared Value: the role of business in society. In: Nestlé S.A., The Nestlé “Creating Shared Value” Report 2007 (pp. 2-3). Vevey: Nestlé S.A., Public Affairs.
Davis, M.L. (2008). Chief Executive’s report. In: Xstrata plc, Sustainability Report 2007 (pp. 2-3). Zug: Xstrata plc.
Mulcahy, A.M. (2007). Chairman’s letter. In: Xerox Corporation, 2007 Report on Global Citizenship. Our Word. Our Work. Our World (pp. 1-3). Norwalk: Xerox Corporation.
Nienaber, P.J (2008). CEO’s review. In: Xstrata Alloys, Sustainability Report 2007 (p. 1). Rustenburg: Xstrata South Africa (Pty) Limited Alloys.
PepsiCo, Inc. (2008). Sustainability Q&A with Indra Nooyi, PepsiCo Chairman and CEO. In: PepsiCo, Inc., Overview2007: Our Sustainability Journey (pp. 2-3). Purchase: PepsiCo, Inc.
Royal Philips Electronics (2008). Interview with the president. In: Royal Philips Electronics, Sustainability Report 2007: Simpler, Stronger, Greener (pp. 8 -11). Amsterdam: Koninklijke Philips Electronics N.V.
Royal Philips Electronics (2008). Becoming simpler, stronger and greener. In: Royal Philips Electronics, Sustainability Report 2007: Simpler, Stronger, Greener (pp. 12-13). Amsterdam: Koninklijke Philips Electronics N.V.
Sarin, A. (2008). Message from the Chief Executive. In: Vodafone Group Plc, Corporate Responsibility Report for the year ended 31March 2008: One Strategy (pp. 9-10). Newbury: Vodafone Group Plc.
Su, S. (2008). Welcome from the Chairman. In: Sinopec Corp. – China Petroleum & Chemical Corporation, Sustainable Development Report 2007 (pp. 3-4). Beijing: Sinopec Corp. – China Petroleum & Chemical Corporation.
Wijers, H. (2008). Chairman’s statement. In: Akzo Nobel N.V., Sustainability Report 2007: Year of Transformation (pp. 8-9). Amsterdam: Akzo Nobel Corporate Communications.

What I want to do in this essay is to discuss the notion of meta-argumentation by summarizing some past work and motivating a future investigation (which, for obvious reasons, I shall label the “Dutch” project). The discussion is meant to make a plea partly for the theoretical and methodological importance and fruitfulness of meta-argumentation in general, and partly for approaching from the viewpoint of meta-argumentation a particular (Dutch-related) topic that is especially relevant on the present occasion for reasons other than methodology and theory. I hope that the potential appeal of this aspect of the essay – combining methodological orientation and theoretical conceptualization with empirical and historical content – will make up for whatever shortcomings it may possess from the point of view of substantive detail about, and completed attainment of, the Dutch project.

1. Historical Context of William the Silent’s Apologia (1581)
In May 1581, the States-General of the Low Countries met here[i] in Amsterdam to draft a declaration of independence from Philip II, King of Spain, who had ruled this region since 1555. In the course of the summer, this congress moved to The Hague, where the declaration was concluded at the end of July. This declaration is called the “act of abjuration”, meaning that these provinces were thereby abjuring their allegiance to the King of Spain.[ii]

This act of abjuration was taking place in the midst of an armed conflict that had already lasted twenty-five years and was to continue for another quarter century. The conflict was partly a war of national independence for the modern Netherlands. However, the conflict was also a civil war within the Low Countries stemming from religious and ethnic differences: the main religious difference was between Catholics and Protestants, while the main ethnic difference was between Dutch-speaking northerners and French-speaking Walloons in the south; eventually this civil war was partially, although not completely, resolved by the split between Belgium and The Netherlands. Finally, the conflict was partly a democratic revolution, in which the people were objecting to taxation without representation and defending local rights vis-à-vis centralized government.

The act of abjuration was occasioned by a proclamation issued the previous year by King Philip against the leader of the revolt, William of Nassau, Prince of Orange, now known as William the Silent. Philip’s proclamation banned William from the Low Countries and called for his arrest or assassination, promising the assassin a large sum, a title of nobility, and a pardon for any previous crimes.

William was the most important leader of the revolt, popular among the nobility as well as common people, influential among Catholics as well as Protestants, and fluent in both French and Dutch. He was becoming increasingly effective in his leadership, especially in the provinces of Holland and Zealand, which were more independent-minded than the other fifteen. Although the difficulty of the struggle and his assassination four years later prevented him from seeing his efforts come to fruition, he paved the way for the later success. For even after his death his qualities could serve as a model: he was usually regarded as thoughtful, prudent, moderate, tolerant, and politically astute and skillful.

William had been the first-born, in 1533, to the Protestant Count of Nassau, in Germany. At age eleven, he inherited from a cousin vast possessions in the Low Countries and elsewhere, including the small principality of Orange in France and the title of Prince. This inheritance was approved on one condition by Charles V, Holy Roman Emperor, King of Spain, and father of Philip II: that William’s parents relinquish their parental authority. Thus, he was thereafter educated as a French-speaking and Dutch-speaking Catholic in the Low Countries. Later, however, in 1573, he re-joined the Reformed Church, while continuing to uphold as supreme the right of freedom of conscience.

In response to Philip’s proclamation, William produced a document entitled Apologia (William 1581; 1858; 1969). This was presented to the States-General in December 1580. The following year it was published as a booklet of one hundred pages in the original French version, as well as in English, Dutch, German, and Latin translations. Copies were sent to all rulers of Christendom.

Thus, in the years 1580-1581, in the context of the ongoing armed conflict in the Low Countries, the Netherlands revolt produced a remarkable triad of documents: a proclamation of proscription and assassination by King Philip II of Spain against William of Orange; a defense by William from Philip’s accusations; and a declaration of independence from Philip’s sovereignty by the States-General of the Low Countries. Of these documents, William’s Apologia is the most informative, because it is the longest, because it summarizes Philip’s charges, and because it anticipates the declaration of independence. It is not surprising that the Apologia went through sixteen editions in the following two decades (Wansink 1969, p. vii).

William’s Apologia is also a more argumentative text than the other two. It is an intense piece of argumentation, for it attempts to do several things: to refute Philip’s accusations; to advance countercharges; to justify William’s own behavior; and to justify the right of the Low Countries to independence.

This judgment about the argumentational import of William’s Apologia is widely shared. For example, Voltaire described it as one of the most beautiful arguments in history.[iii] The nineteenth-century American historian John Motley expressed the following judgment: William “possessed a ready eloquence – sometimes impassioned, oftener argumentative, always rational. His influence over his audience was unexampled in the annals of that country or age, yet he never condescended to flatter the people” (Motley 1883, vol. 3, p. 621); and Motley was the author of a monumental history of the Netherlands revolt, in seven volumes, totaling 3400 pages (Motley 1856; 1860). Even a more critical historian, himself a Dutchman, who was the dean of twentieth-century scholars of Dutch history, Pieter Geyl, judged the following: William of “Orange’s greatness as a leader of the Netherlands people lay precisely in his unsurpassed talent for co-operating with the States assemblies … Persuasion was what he excelled in” (Geyl 1958, p. 193). Finally, in the past decade William’s Apologia has attracted the attention of Frans van Eemeren and Peter Houtlosser (1999; 2000; 2003), who have examined it from the point of view of the pragma-dialectical theory of argumentation. In fact, I can report that it was their articles that first awakened my interest in this text. Their judgment, added to that of Voltaire, Motley, and Geyl, and my earlier historical considerations, suggest that William’s Apologia is a candidate for analysis on the present occasion.

2. Universal Cultural Significance of William’s Apologia
Nevertheless, I hesitate to undertake an analysis of this work. For I am sensitive to the potential criticism that it is risky, rash, or arrogant for an outsider like myself who lives about 10,000 kilometers from The Netherlands to rummage through local history and expect to find anything new or insightful to tell locals (or other interested parties). It’s as if a visitor were to lecture at my University of Nevada, Las Vegas, and pretend to give locals lessons about gambling, hotel administration, or popular entertainment.
On the other hand, an analysis of William’s Apologia may be worthwhile for other reasons, above and beyond the ad hoc, localistic, or antiquarian considerations advanced so far. These additional reasons are philosophical or general-cultural, as well as methodological or epistemological.

The main cultural reason is that William’s Apologia, and the Netherlands revolt which it epitomizes, are of universal significance, and not merely historical curiosities of interest to people who happen to descend from those protagonists.
For example, I have already mentioned that a crucial issue over which William fought was freedom of religion and of individual conscience. Now, let me simply add the obvious, namely that this cluster of freedoms and individual rights is one of the great achievements of modernity, and that it certainly is not going to be superseded by anything which so-called post-modernists have proposed or are going to propose. To be sure, this freedom is subject to abuse, misuse, and atrophy from non-use, as well as perversion and subversion, and so it must be constantly safeguarded and requires eternal vigilance. But these caveats too are a lesson that can be learned from the Netherlands revolt. In fact, in that period, it often happened that, once the Calvinist Protestants got the upper hand in a town or province, they had the tendency to reserve that freedom only for themselves and deny it to the Catholics. However, in William we have someone who defended the legitimate rights of both sides, and opposed the abuses of both.

A second example is provided by the similarities between the 1581 act of abjuration and the American Declaration of Independence of 1776. The similarities center on the political right of the governed to give or withhold their consent to the governors. That is, the Netherlands declaration antedates by about two centuries the American declaration, and thus must be regarded as one of the founding documents in the history of political democracy. And again, needless to say, the same caveats apply to the democratic ideal that apply to the ideal of religion liberty.

Let me conclude these considerations on the universal significance of the Netherlands revolt and William’s Apologia with some quotations from the works of John Motley, the nineteenth-century American mentioned earlier as the author of a monumental history of the revolt. For the eloquence and inspired zeal of this outsider are themselves eloquent and inspiring testimony of that universality.

Motley’s book begins with these words: “The rise of the Dutch Republic must ever be regarded as one of the leading events of modern times … [It was] an organized protest against ecclesiastical tyranny and universal empire … [For] the splendid empire of Charles the Fifth was erected upon the grave of liberty. It is a consolation to those who have hope in humanity to watch, under the reign of his successor, the gradual but triumphant resurrection of the spirit over which the sepulchre had so long been sealed” (Motley 1883, vol. 1, p. iii).
Here, Motley is attributing to the Netherlands revolt two merits, namely its contribution to the ideals of religious freedom and national liberation. But next he speaks of a third merit, which is an epoch-making contribution to the art of politics: “To the Dutch Republic … is the world indebted for practical instruction in that great science of political equilibrium which must always become more and more important as the various states of the civilized world are pressed more closely together …  Courage and skill in political and military combinations enabled William the Silent to overcome the most powerful and unscrupulous monarch of his age” (Motley 1883, vol. 1, pp. iii-iv).

3. The Historical-Textual Approach to Argumentation
So much for the universal significance of William’s Apologia, providing a cultural reason for undertaking an analysis of its argumentation. Now, I go on to the methodological considerations. These are really more pertinent, and it is they that have made me overcome my hesitation in tackling a subject that is apparently so distant from my scholarly concerns.

For a number of years, I have advocated an empirical approach to the study of argumentation which I call the historical-textual approach (Finocchiaro 1980, pp. 256-307; 2005, pp. 21-91). In this approach, the working definition – indeed almost an operational definition – of argumentation is that it occurs typically in written or oral discourse containing a high incidence of illative terms such as: therefore, so, thus, hence, consequently, because, and since.

Here, I contrast the empirical primarily to the apriorist approach, an example of the latter being formal deductive logic insofar as it is regarded as a theory of argument. On the other hand, I do not mean to contrast the empirical to the normative, for the aim of the historical-textual approach is the formulation of normative and evaluative principles besides descriptive, analytical, and explanatory ones. Another proviso is that my empirical approach ought not to be regarded as empiricist, namely as pretending that it can study argumentation with a tabula rasa.

This historical-textual approach is my own variation on the approaches advocated by several scholars. They have other labels, different nuances, and partly dissimilar motivations and aims. Nevertheless, my approach derives partly from that of Michael Scriven and his probative logic; Stephen Toulmin and his methodological approach, as distinct from his substantive model of argument; Henry Johnstone Jr. and his combination of philosophy and rhetoric; and Else Barth and her empirical logic.[iv] Moreover, my approach overlaps with that of Ralph Johnson, Tony Blair, and informal logic; Alec Fisher and his logic of real arguments; and Trudy Govier and her philosophy of argument, meaning real or realistic arguments.[v]

Typically, the historical-textual approach involves the selection of some important text of the past, containing a suitably wide range and intense degree of argumentation. Many of the classics fulfill this requirement, for example, Plato’s Republic, Thomas Aquinas’s Summa Theologica, The Federalist Papers by Alexander Hamilton, John Jay, and James Madison, and Charles Darwin’s Origin of Species. Not all classics would be appropriate: some for lack of argumentation, some for insufficient intensity, and some for insufficient variety. In some cases works other than the classics would serve the purpose, for example collections of judicial opinions by the United States Supreme Court or the World Court in The Hague.

Given this sketch of the historical-textual approach, together with my earlier remarks about William’s Apologia, now perhaps you can begin to see the connection, that is, a possible methodological motivation for undertaking an analysis of that work. But this is just the beginning, and I am not sure that what I have said so far would provide a sufficient motivation for me. So let me go on with my methodological justification.

Following such an historical-textual approach, many years ago I undertook a study of Galileo Galilei’s book, Dialogue on the Two Chief World Systems, Ptolemaic and Copernican. This book is not only the mature synthesis of astronomy, physics, and methodology by the father of modern science, but also the work that triggered Galileo’s Inquisition trial and condemnation as a suspected heretic in 1633; it is also full of arguments for and against the motion of the earth. My study led me to a number of theoretical claims (Finocchiaro 1980, pp. 311-431; 1997, pp. 309-72; 2005, pp. 34-91, 109-80).

For example, the so-called fallacies are typically either non-fallacious arguments, or non-arguments, or inaccurate reconstructions of the originals; but many arguments can be criticized as fallacious in various identifiable ways. There are important asymmetries between the positive and the negative evaluation of arguments, although one particular alleged asymmetry seems untenable, namely the allegation that it is possible to prove formal validity but not formal invalidity. One of the most effective ways of criticizing arguments is to engage in ad hominem argumentation in the seventeenth century meaning of this term, namely to derive a conclusion unacceptable to opponents from premises accepted by them (but not necessarily by the arguer). Finally, argumentation plays an important and still under-studied and unappreciated role in science.

4. The Meta-argumentation Project
All this may be new to some of you, familiar to a few others, but almost ancient history to me. For more recently, I have been focusing on meta-argumentation. It’s not that I have abandoned my historical-textual approach, but that I have found it fruitful to apply it to a special class of arguments, called meta-arguments. On this subject, I want to acknowledge Erik Krabbe (1995; 2002; 2003) as a source of inspiration and encouragement. Paraphrasing his definition of metadialogue, I define a meta-argument as an argument about one or more arguments. A meta-argument is contrasted to a ground-level argument, which is typically about such topics as natural phenomena, human actions, or historical events.

Meta-arguments are special in at least two ways, in the sense of being crucially important to argumentation theory, and in the sense of being a particular case of argumentation. First, meta-arguments are crucially important because argumentation theory consists, or ought to consist, essentially of meta-argumentation; thus, studying the meta-arguments of argumentation theorists is a meta-theoretical exercise in the methodology of our discipline. Second, meta-arguments as just defined are a particular case of argument-tation, and so their study is or ought to be a particular branch of argumentation theory.

Consequently, my current project has two main parts. In both, because of the historical-textual approach, the meta-arguments under investigation are real, realistic, or actual instances of argumentation. But in the meta-theoretical part, the focus is on important arguments from recent argumentation theory. In the other part, the focus is on famous meta-arguments from the history of thought.

Before illustrating this project further, let me elaborate an immediate connection with William’s Apologia. In fact, William’s text is not just an intense and varied piece of argumentation, as mentioned before, but it is also a meta-argument since it is primarily a response to King Philip’s proclamation. But Philip’s proclamation gave reasons why William should be proscribed and assassinated, and however logically incoherent and mean-spirited those reasons may have been, they constitute an argument, at least for those of us who uphold the fundamental distinction between an argument and a good argument. On the other hand, Philip’s proclamation is a ground-level argument, and the same is true of the States-General’s act of abjuration. Thus, my motivation for undertaking an analysis of William’s Apologia can now be fleshed out further. I can go beyond my earlier remark that it is a candidate for study by argumentation scholars because it is a famous example of intense and varied argumentation; now I can add that the text is a good candidate for analysis in a study of meta-argumentation conducted in accordance with the historical-textual approach.

However, how promising is such a project? I must confess that the stated motivation, even with the addition just made, would still be insufficient, at least for me, if this were my first study of a famous meta-argument in terms of the historical-textual approach; that is, if I had not already conducted some such studies and obtained some encouraging results. Moreover, it is important that this project plans to study famous meta-arguments in conjunction with currently important theoretical arguments because, as mentioned earlier, the hope is not merely to contribute to a particular branch of argumentation studies, however legitimate that may be, but also to address some key issues of argumentation theory in general. Thus, I need to at least summarize some of my previous meta-argumentative studies, in order to strengthen my methodological plea for an analysis of William’s Apologia.

5. Meta-argumentation in the Subsequent Galileo Affair
Let me begin by saying a few words about one of my previous studies of meta-argumentation (Finocchiaro 2010) that is intermediate between my current project and my earlier study of the ground-level arguments in Galileo’s Dialogue. At a subsequent stage of my research, I discovered a related set of significant arguments that are primarily meta-arguments. Their existence was not as easily detectable, because they are not found within the covers of a single book, and because initially they do not appear to focus on a single issue. This discovery required a laborious work of historical interpretation, philosophical evaluation, and argument reconstruction.

I am referring to the arguments that make up the subsequent Galileo affair, as distinct from the original affair. By the original Galileo affair I mean the controversy over the earth’s motion that climaxed with the Inquisition’s condemnation of Galileo in 1633. By the subsequent affair I mean the ongoing controversy over the rightness of Galileo’s condemnation that began then and continues to our own day. The arguments that define the original affair (and that are primarily ground-level) are relatively easy to find, the best place being, as mentioned, Galileo’s own book. On the other hand, the arguments that make up the subsequent affair (and that are primarily meta-arguments) must be distilled out of the commentaries on the original trial produced in the past four centuries by all kinds of writers: astronomers, physicists, theologians, churchmen, historians, philosophers, cultural critics, playwrights, novelists, and journalists.

Let me give you some examples, both to give you an idea of the substantive issues of the subsequent affair and of the fact that it consists of meta-arguments. To justify the claim that the Inquisition was right to condemn Galileo, the following reasons, among others, have been given at various times by various authors (see Finocchiaro 2010, pp. xx-xxxvii, 155-228). (1) Galileo failed to conclusively prove the earth’s motion, which was not accomplished until Newton’s gravitation (1687), Bradley’s stellar aberration (1729), Bessel’s annual stellar parallax (1838), or Foucault’s pendulum (1851). (2) Galileo was indeed right that the earth moves, but his supporting reasons, arguments, and evidence were wrong, ranging from the logically invalid and scientifically incorrect to the fallacious and sophistical; for example, his argument based on a geokinetic explanation of the tides is incorrect. (3) Galileo was indeed right to reject the scientific authority of Scripture, but his supporting reasoning was incoherent, and his interference into theology and scriptural interpretation was inappropriate. (4) Galileo may have been right scientifically (earth moves), theologically (Scripture is not a scientific authority), and logically (reasoning), but was wrong legally; that is, he was guilty of disobeying the Church’s admonition not to defend earth’s motion, namely not to engage in argumentation, or at least not to evaluate the arguments on the two sides of the controversy.

After such meta-arguments are found and reconstructed, one must evaluate them. In accordance with my historical-textual approach, part of the evaluation task involves reconstructing how such arguments have been assessed in the past four centuries. But I also had another idea. One could try to identify the essential elements of the approach which Galileo himself followed in the original controversy over the earth’s motion, and then adapt that approach to the subsequent controversy. This turned out to be a fruitful idea.

In particular, two principles preached and practiced by Galileo were especially relevant. Influenced by the literature on informal logic, I label them the principles of open-mindedness and fair-mindedness, but here I am essentially paraphrasing his formulations. Open-mindedness is the willingness and ability to know and understand the arguments against one’s own claims. Fair-mindedness is the willingness and ability to appreciate and strengthen the opposing arguments before refuting them.

Thus, I was led to the following overarching thesis about the meta-arguments making up the subsequent Galileo affair: that is, the anti-Galilean arguments can and should be successfully criticized by following the approach which Galileo himself used in criticizing the anti-Copernican arguments, and this is an approach characterized by open-mindedness and fair-mindedness. In short, at the level of interpretation, I argue that the subsequent Galileo affair can be viewed as a series of meta-arguments about the pro- and anti-Copernican ground-level arguments of the original affair; at the level of evaluation, I argue that today, in the context of the Galileo affair and the controversies over the relationship between science and religion and between institutional authority and individual freedom, the proper defense of Galileo should have the reasoned, critical, open-minded, and fair-minded character which his own defense of Copernicanism had.

6. Theoretical Meta-arguments
Let us now go on to my current project studying meta-argumentation in an historical-textual manner. I begin with some examples of the meta-theoretical part of this project.[vi]
One of these meta-arguments is Ralph Johnson’s justification of his dialectical definition of argument (cf. Finocchiaro 2005, pp. 292-328). I start with a contrast between the illative and the dialectical definitions, but distinguish three versions of the latter: a moderate conception for which the dialectical tier is sufficient but not necessary; a strong conception for which the dialectical tier is necessary but not sufficient; and an hyper conception for which the dialectical tier is necessary and sufficient. Johnson’s conclusion is the strongly dialectical conception. His argument contains an illative tier of three supporting reasons, and a dialectical tier consisting of four criticisms of the illative conception and replies to six objections. The result of my analysis is the conclusion that the moderate conception is correct, namely, that an argument is an attempt to justify a conclusion by either supporting it with reasons, or defending it from objections, or both. My argument contains supporting reasons appropriated from the acceptable parts of Johnson’s argument, and criticism of his strong conception. I also defend my moderate conception from some objections.

Another example involves the justification of the hyper dialectical definition of argument advanced by Frans van Eemeren and the pragma-dialectical school (cf. Finocchiaro 2006). The hyper dialectical definition of argument claims that an argument is simply a defense of a claim from objections. Their meta-argument is difficult to identify, but it can be reconstructed. Before criticizing it, I defend it from one possible criticism, but later I argue that it faces the insuperable objection that the various analyses which pragma-dialectical theorists advance to support their definition do not show it is preferable to all alternatives. Then I advance an alternative general argument for the unique superiority of the hyper definition over the others, but apparently it fails because of the symmetry between supporting reasons and replies to objections. My conclusion is that the moderately dialectical conception is also preferable to the hyper dialectical definition.

Next, I have examined the arguments for various methods of formal criticism by Erik Krabbe, Trudy Govier, and John Woods (cf. Finocchiaro 2007a). This turned out to be primarily a constructive, analytical, or reconstructive exercise, rather than critical or negative. Krabbe (1995) had shown that formal-fallacy criticism (and more generally, fallacy criticism) consists of metadialogues, and that such metadialogues can be profiled in ways that lead to their proper termination or resolution. I reconstruct Krabbe’s metadialogical account into monolectical, meta-argumentative terminology by describing three-types of meta-arguments corresponding to the three ways of proving formal invalidity which he studied: the trivial logic-indifferent method, the method of counterexample situation, and the method of formal paraphrase. A fourth type of meta-argument corresponds to what Govier (1985) calls refutation by logical analogy. A fifth type of meta-argument represents my reconstruction of arguments by parity of reasoning studied by Woods and Hudak (1989).

Another example is provided by the meta-arguments about deep disagreements. Here, I examine the arguments advanced by such scholars as Robert Fogelin, John Woods, and Henry Johnstone, Jr., about what they variously call deep disagreements, intractable quarrels, standoffs of force five, and fundamental philosophical controversies (see Fogelin 1985, 2005; Woods 1992, 1996; Johnstone 1959, 1978). As much as possible their views, and the critiques of them advanced by other scholars, are reconstructed as meta-arguments. From my analysis, it emerges that deep disagreements are rationally resolvable to a greater degree than usually believed, but that this can be done only by the use of such principles and practices as the following: the art of moderation and compromise (codified as Ramsey’s Maxim); open-mindedness; fair-mindedness; complex argumentation; meta-argumentation; and ad hominem argumentation in a sense elaborated by Johnstone and corresponding to the seventeenth-century meaning, mentioned earlier.

Finally, another fruitful case study has dealt with conductive meta-arguments. The term “conductive” argument was introduced by Carl Wellman (1971), as a third type of argumentation besides deduction and induction. In this context, a conductive argument is primarily one in which the conclusion is reached nonconclusively based on more than one separately relevant supporting reason in favor and with an awareness of at least one reason against it. Conductive arguments are more commonly labeled pro-and-con arguments, or balance-of-considerations arguments. They are ubiquitous, especially when one is justifying evaluations, recommendations, interpretations, or classifications. Here I reconstruct Wellman’s original argument, the constructive follow-up arguments by Govier (1980; 1987, pp. 55-80; 1999, pp. 155-80) and David Hitchcock (1980; 1981; 1983, pp. 50-53, 130-34; 1994), and the critical arguments by Derek Allen (1990; 1993) and Robert Ennis (2001; 2004). My own conclusion from this analysis is that so-called conductive arguments are good examples of meta-arguments; for a crucial premise of such arguments is a balance-of-considerations claim to the effect that the reasons in favor of the conclusion outweigh the reasons against it; such a claim can be implicit or explicit; but to justify it one needs a subargument which is a meta-argument; hence, while the conclusion of a conductive argument is apparently a ground-level proposition, a crucial part of the argument is a meta-argument.

7. Famous Meta-arguments
These examples should suffice as a summary of the meta-theoretical part of my study of meta-argumentation in accordance with the historical-textual approach. The other part was a study of famous meta-arguments that are important for historical or cultural reasons. Obviously, the meta-arguments in William’s Apologia are of the latter sort. So it will be useful to look at what some of these previous studies have revealed.

A striking example is provided by chapter 2 of John Stuart Mill’s essay On Liberty (cf. Finocchiaro 2007c). It can be reconstructed as a long and complex argument for freedom of discussion. The argument consists of three subarguments, each possessing illative and dialectical components. The illative component is this. Freedom of discussion is desirable because, first, it enables us to determine whether an opinion is true; second, it improves our understanding and appreciation of the supporting reasons of true opinions, and of their practical or emotional meaning; and third, it enables us to understand and appreciate every side of the truth, given that opinions tend to be partly true and partly false and people tend to be one-sided. The dialectical component consists of replies to ten objections, five in the first subargument, three in the second, one in the third, and one general.

So reconstructed, Mill’s argument is a meta-argument, indeed it happens to be also a contribution to argumentation theory. For its main conclusion can be rephrased as the theoretical claim that freedom of argument is desirable. A key premise, which Mill assumes but does not support, turns out to be the moderately dialectical conception of argument. And one of his principal claims is the thesis that argumentation is a key method in the search for truth.

Another famous meta-argument occurs in Mill’s book on The Subjection of Women (cf. Finocchiaro 2007b). The whole book is a ground-level argument for the thesis that the subjection of women is wrong and should be replaced by liberation and equality. The meta-argument is found in the first part of chapter 1. Then in the rest of that chapter, he replies to a key objection to his own thesis. Finally, in the other three chapters he articulates three reasons supporting that thesis. Mill begins by formulating the problem that the subjection of women is apparently a topic where argumentation is counterproductive or superfluous. He replies by rejecting the principle of argumentation that generates this problem and replacing it by a more nuanced principle. However, this principle places on him the burden of causally undermining the universal belief in the subjection of women, to pave the way for argumentation on the merits of the issue. Accordingly, he argues that the subjection of women derives from the law of the strongest, but that this law is logically unsound and morally questionable, and hence that custom and feeling provide no presumption in favor of the subjection of women. Additionally, Mill thinks that in this case he can make a predictive extrapolation; accordingly, he argues that there is a presumption against subjection based on the principle of individual freedom. This predictive extrapolation and the causal undermining are complementary meta-arguments.

Now, these two meta-arguments may also be viewed, respectively, as the criticism of an objection, and the statement of a supporting reason, and hence as elements of the dialectical and illative tiers, rather than as a distinct meta-argumentative part of the overall argument. This possibility raises the theoretical issue that there may be a symmetry between meta and ground levels analogous to the symmetry between illative and dialectical tiers; if so, then meta-argumentation would be not only an explicit special type of argument, but also an implicit aspect of all argumentation,[vii] distinct from but related to the illative and dialectical components.

A third example of famous meta-argumentation is the critique of the theological design argument found in David Hume’s Dialogues concerning Natural Religion (cf. Finocchiaro 2009). Hume’s critique is a complex meta-argument, consisting of two main parts, one interpretive, the other critical. His interpretive meta-argument claims that the design argument is an inductive ground-level argument, with a complex structure, consisting of three premises and two sub-arguments, one of which sub-arguments is an inductive generalization, while the other is a statistical syllogism. Hume’s critical meta-argument argues that the design argument is weak because two of its three premises are justified by inadequate sub-arguments; because its main inference embodies four flaws; and because the conclusion is in itself problematic for four reasons. Finally, he also argues that the design argument is indirectly undermined by two powerful ground-level arguments, involving the problem of evil; they justify conclusions that are in presumptive tension with the conclusion of the design argument, while admittedly not in strict contradiction with it.

Here, the main theoretical implication is along the following lines. Hume’s critique embodies considerable complexity, so much so that it could be confusing. However, such complexity becomes quite manageable in a meta-argumentation approach; this means that the concept of meta-argument can serve as a principle of simplification, enhancing intelligibility, but without lapsing into over-simplification.

8. Conclusion
In summary, (F) the analysis of William the Silent’s Apologia is a very promising project in argumentation studies, for two reasons, a general one involving my historical-textual approach, and a more specific and important one involving my meta-argumentation project.

First, generally speaking, (Fa11) this work contains argumentation that is intense and varied, as revealed by (Fa111) even a cursory reading, as well as (Fa112) the considered judgment of many authorities. Moreover, (Fa12) the issues it discusses are universally significant because they involve (Fa121) freedom of religion, (Fa122) the right to national independence, (Fa123) the ideal of democratic consent, and (Fa124) the art of political equilibrium. Thus, (Fa1) this text is susceptible of being analyzed in accordance with the historical-textual approach to argumentation in general. But we have seen that (Fa2) the historical-textual approach is fruitful; for example, (Fa21) it has yielded interesting results by studying the arguments about the motion of the earth in Galileo’s Dialogue.

More specifically and more importantly, (Fb1) William’s Apologia is a piece of meta-argumentation since (Fb11) it is a response to a proclamation that is itself an argument. But we have seen that (Fb2) the historical-textual study of meta-arguments is proving to be a fruitful project. For example, (Fb21) it has already yielded some results with regard to the meta-arguments that constitute the subsequent Galileo affair. More to the point, (Fb22) it is yielding interesting results with regard to the meta-arguments of leading argumentation theorists, dealing with topics such as (Fb221) the strongly dialectical concept of argument, (Fb222) the hyper dialectical concept of argument, (Fb223) methods of formal criticism, (Fb224) deep disagreements, and (Fb225) conductive arguments; and (Fb23) it is also yielding interesting results with regard to famous meta-arguments, such as Mill on (Fb231) liberty of argument and on (Fb232) women’s liberation, and (Fb233) Hume on the theological design argument.

What I have just summarized is (dare I say it?) my argument, such as it is, in this address here today; that is, the reasons why I think it would be fruitful to analyze William’s Apologia from the point of view of meta-argumentation and the historical-textual approach; that is, my prolegomena to a future meta-argumentative and historical-textual study of this Dutch classic.

If I had more time, I might discuss the details of the propositional macrostructure of my argument, as you can visualize in the following diagram:[viii]

This would reinforce the fact that, after all, I have been arguing for the past hour, however modestly in intention, execution, and results. Could I have done anything less? Or different? I suppose I could have described the details of William’s meta-argumentation, which of course I am now committed to doing sooner or later. But this description, even without motivation or justification, would have taken the whole hour. Moreover, my describing by itself would not have been an actual instantiation of argumentation, let alone meta-argumentation. On the contrary, in this address I wanted, among other things, to practice what I preached.

NOTES
[i] A slightly shorter version of this paper was delivered as a keynote address to the Seventh Conference of the International Society for the Study of Argumentation at the University of Amsterdam, on 30 June 2010. This venue accounts for my choice of this word here, as well as for the similar self-referential remarks in the last two paragraphs in section 8 below.
[ii] This episode is discussed in Motley 1883, vol. 3, pp. 507-9; Wedgwood 1944, p. 222; Geyl 1958, pp. 183-84; and Swart 1978, p. 35. My account in the rest of this paper is also based on these works, but from here on no specific references will usually be given, except for quotations and a few other specific items.
[iii] Quoted in Eemeren and Houtlosser 2003, p. 178. I am paraphrasing, for Voltaire said monument, which I am reading as argument because the “monument” we are dealing with is linguistic rather than physical. Motley (1883, vol. 3, p. 493) paraphrases monument as document.
[iv] See Scriven (1976; 1987) and cf. Finocchiaro 2005, pp. 5-7; see Toulmin 1958 and cf. Finocchiaro (1980, pp. 303-305; 2005, pp. 6-7); see Johnstone (1959; 1978) and cf. Finocchiaro (2005, pp. 277-91, 329-39); see Barth 1985, Barth and Krabbe 1992, Barth and Martens 1982, Krabbe et al. 1993, and cf. Finocchiaro (2005, pp. 46-64, 207-10).
[v] See Blair and Johnson 1980, Johnson 1987, Johnson and Blair 1994, and cf. Finocchiaro (2005, pp. 21-33); and see Fisher (1988; 2004) and Govier (1987; 1999; 2000, pp. 289-90), and cf. Finocchiaro (2005, pp. 1-105, 329-429).
[vi] One of the referees raised an objection to this part of the project along the following lines: in order to assess the arguments that make up a given argumentation theory, one has to use either the evaluation criteria of the same theory or those of another theory; but if one uses the same criteria, it is not obvious that such self-reflective exercise is possible or fair (the latter because it might automatically yield a favorable assessment); on the other hand, if one uses the evaluation criteria of another theory, then it is also not obvious that such an external evaluation is possible or fair (the latter because it might automatically yield an unfavorable assessment); therefore, this meta-theoretical project is doomed from the start since it may very well be impossible or unfair.
My reply is that this objection seems to assume uncritically a relationship between the theory and the practice of argumentation that may be the reverse of the right one. My inclination is practically oriented, in the sense of giving primacy to the practice of meta-argumentation. That is: let us try to do the meta-theoretical exercise; if it can be done, that shows that it is possible; moreover, let us try to be fair-mined in doing it; if we succeed in doing it fairly, that shows that the meta-theoretical evaluation can be fair; thus, let us postpone questions of possibility and fairness until afterwards. Moreover, the objection perhaps proves too much, in the sense that if what it says about evaluation or assessment were correct, then it would be likely to apply also to interpretation or reconstruction, in which case it would be suggesting that theoretical meta-arguments perhaps cannot even be understood, at least not from an external point of view; and such a parallel objection strikes me as being a reductio ad absurdum of its own assumptions.
[vii] As one of the referees pointed out, this hypothesis may be viewed as a special case of a thesis widely held in communication studies. For example, Bateson (1972, pp. 177-78) has claimed that “human verbal communication can operate and always does operate at many contrasting levels of abstraction. These range in two different directions … metalinguistic … [and] metacommunicative.” Similarly, Verschueren (1999, p. 195) has maintained that “all verbal communication is self-referential to a certain degree … all language use involves a constant interplay between pragmatic and metapragmatic functioning … reflexive awareness is at the very core of what happens when people use language.”
I take this coincidence or correspondence as an encouraging sign, but I think it would be a mistake to exploit it for confirmatory purposes. In particular, such general theses cannot be used to justify my particular hypothesis about meta-argumentation because they are formulated and defended in a context and with evidence that does not involve the phenomenon of argumentation, but rather other linguistic and communicative practices. For example, Bateson (1972, pp. 177-93) is dealing with such phenomena as playing, threats, histrionics, rituals, psychotherapy, and schizophrenia; and of Verschueren’s (1999, pp. 179-97 ) fifty-four examples of metapragmatic use of language, only two involve (simple, ground-level) arguments. Thus I feel they have not established that their generalizations apply to argumentative communication, and the question whether this particular application holds is the same question whether my meta-argumentation hypothesis is correct.  Moreover, I would stress that both authors (Bateson 1972, p. 178; Verschueren 1999, pp. 183-87) are keen to point out that the metalevel aspect of the phenomena they study is a matter of degree and is usually implicit; on the other hand, my own meta-argumentation project focuses on very explicit cases.
The same referee also pointed out the other side of the coin of this potential confirmation of my hypothesis by the widely held generalization from communication studies. That is, perhaps my distinction between ground-level and meta-argumentation, together with my hypothesis about the implicitly meta-argumentative aspect of all argumentation, is afflicted by the difficulties stemming from the self-referential paradoxes such as Russell’s and the liar’s paradox. For example, Bateson (1972, pp. 179-80) is worried that when two humans or animals are playing by simulating a physical combat, the meta-communicative “message ‘This is play’ … contains those elements which necessarily generate a paradox of the Russellian or Epimenides type – a negative statement containing an implicit negative metastatement. Expanded, the statement ‘This is play’ looks something like this: … ‘These actions in which we now engage, do not denote what would be denoted by those actions which these actions denote’.” Recall that Russell’s paradox exposes the self-contradiction of the notion of a set of all sets that are not members of themselves, and that the liar’s paradox is the self-contradiction of the statement that this statement is false.
However, my reply to this potentially negative criticism is analogous to my reply to the earlier potentially strengthening confirmation. I see the difficulty with the Russellian set and with the liar’s sentence, and I see some similarity between them and Bateson’s meta-communicative message that “this fighting is play”; but I see no similarity with my notion of a meta-argument, its distinction from a ground-level argument, and their relationship; and until and unless a similar paradox is specifically derived regarding meta-argumentation, I shall not worry.
[viii] For an explanation of such diagrams, which are now common in the literature and come in various slightly different versions, see, for example, Scriven (1976, pp. 41-43) and Finocchiaro (1980, pp. 311-31; 1997, pp. 309-35; 2005, pp. 39-41).

REFERENCES
Allen, D. (1990). Trudy Govier’s Problems in Argument Analysis and Evaluation. Informal Logic, 12, 43-62.
Allen, D. (1993). Relevance, conduction, and Canada’s rape-shield law. Informal Logic, 15, 105-22.
Barth, E.M. (1985). A new field: Empirical logic. Synthese, 63, 375-88.
Barth, E.M., & Martens, J.L. (Eds.). (1982). Argumentation: Approaches to Theory Formation. Amsterdam: John Benjamins.
Bateson, G. (1972). Steps to an Ecology of Mind. New York: Ballantine Books.
Blair, J.A., & Johnson, R.H. (Eds.). (1980). Informal Logic: The First International Symposium. Inverness: Edgepress.
Eemeren, F.H. van, & Houtlosser, P. (1999). William the Silent’s argumentative discourse. In F.H. van Eemeren, R. Grootendorst, J.A. Blair & C.A. Willard (Eds.), Proceedings of the Fourth International Conference of the International Society for the Study of Argumentation (pp. 168-171). Amsterdam: Sic Sat.  
Eemeren, F.H. van, & Houtlosser, P. (2000). The rhetoric of William the Silent’s Apologie: A dialectical perspective. In T. Suzuki & T. Kata (Eds.), Proceedings of the First Tokyo Conference on Argumentation (pp. 37-40). Tokyo: Japan Debate Association.
Eemeren, F.H. van, & Houtlosser, P. (2003). Strategic manoeuvering: William the Silent’s Apologie, a case in point. In L.I. Komlósi, P. Houtlosser & M. Leezenberg (Eds.), Communication and Culture (pp. 177-185). Amsterdam: Sic Sat.
Ennis, R.H. (2001). Argument appraisal strategy. Informal Logic, 21, 97-140.
Ennis, R.H. (2004). Applying soundness standards to qualified reasoning. Informal Logic, 24, 23-39.
Finocchiaro, M.A. (1980). Galileo and the Art of Reasoning: Rhetorical Foundations of Logic and Scientific Method. (Boston Studies in the Philosophy of Science, vol. 61.)  Dordrecht: Reidel [now Springer].
Finocchiaro, M.A. (Ed. & Trans.). (1997). Galileo on the World Systems: A New Abridged Translation and Guide. Berkeley: University of California Press.
Finocchiaro, M.A. (2005). Arguments about Arguments: Systematic, Critical, and Historical Essays in Logical Theory. New York: Cambridge University Press.
Finocchiaro, M.A. (2006). Reflections on the hyper dialectical definition of argument. In
Finocchiaro, M.A. (2007a). Arguments, meta-arguments, and metadialogues: A reconstruction of Krabbe, Govier, and Woods. Argumentation, 21, 253-68.
Finocchiaro, M.A. (2007b). Famous meta-arguments: Part I, Mill and the tripartite nature of argumentation.” In H.V. Hansen, C.W. Tindale, J.A. Blair, R.H. Johnson & D.M. Godden (Eds.), Dissensus and the Search for Common Ground. Windsor: Ontario Society for the Study of Argumentation. CD-ROM. ISBN 978-0-9683461-5-0.
Finocchiaro, M.A. (2007c). Mill on liberty of argument: A dialectical approach. In H.V. Hansen & R. Pinto (Eds.), Reason Reclaimed (pp. 121-34). Newport News: Vale Press.
Finocchiaro, M.A. (2009). Meta-argumentation in Hume’s critique of the design argument. In J. Ritola (Ed.), Argument Cultures: Proceedings of OSSA 09. Windsor: Ontario Society for the Study of Argumentation. CD-ROM, ISBN 978-0-920233-51-1.
Finocchiaro, M.A. (2010). Defending Copernicus and Galileo: Critical Reasoning in the Two Affairs. (Boston Studies in the Philosophy of Science, vol. 280.) Dordrecht: Springer.
Fisher, A. (1988). The Logic of Real Arguments. Cambridge: Cambridge University Press.
Fisher, A. (2004). The Logic of Real Arguments (2nd ed.). Cambridge: Cambridge University Press.
Fogelin, R.J. (1985). The logic of deep disagreements. Informal Logic, 7, 1-8.
Fogelin, R.J. (2005). The logic of deep disagreements. Informal Logic, 25, 3-11.
Geyl, P. (1958). The Revolt of the Netherlands, 1555-1609 (2nd ed.). London: Ernest Benn.
Govier, T. (1980). Carl Wellman’s Challenge and Response. Informal Logic Newsletter, 2(2), 10-15.
Govier, T. (1985). Logical analogies. Informal Logic, 7, 27-33.
Govier, T. (1987). Problems in Argument Analysis and Evaluation. Dordrecht: Foris.
Govier, T. (1999). The Philosophy of Argument. Newport News: Vale Press.
Govier, T. (2000). Critical review: Johnson’s Manifest Rationality. Informal Logic, 20, 281-91.
Hitchcock, D. (1980). Deductive and inductive types of validity. Informal Logic Newsletter, 2(3), 9-10.
Hitchcock, D. (1981). Deduction, induction, and conduction. Informal Logic Newsletter, 3(2), 7-15.
Hitchcock, D. (1983). Critical Thinking. Toronto: Methuen.
Hitchcock, D. (1994). Validity in conductive arguments. In R.H. Johnson & J.A. Blair (Eds.), New Essays in Informal Logic (pp. 56-66). Windsor: Informal Logic Publications.
Johnson, R.H. (1987). Logic naturalized. In F.H. van Eemeren, R. Grootendorst, J.A. Blair & C.A. Willard (Eds.), Argumentation across the Lines of Discipline: Proceedings of the Conference on Argumentation (pp. 47-56). Dordrecht: Foris.
Johnstone, H.W., Jr. (1959). Philosophy and Argument. University Park: Pennsylvania State University Press.
Johnstone, H.W., Jr. (1978). Validity and Rhetoric in Philosophical Argument. University Park: The Dialogue Press of Man & World.
Krabbe, E.C.W. (1995). Can we ever pin one down to a formal fallacy? In F.H. van Eemeren, R. Grootenodrst, J.A. Blair & C.A. Willard (Eds.), Analysis and Evaluation: Proceedings of the Third ISSA Conference on Argumentation, vol. 2 (pp. 333-344). Amsterdam: Sic Sat.
Krabbe, E.C.W. (2002). Profiles of dialogue as a dialectical tool. In F.H. van Eemeren (Ed.), Advances in Pragma-Dialectics (pp. 153-167). Amsterdam: Sic Sat.
Krabbe, E.C.W. (2003). Metadialogues. In F.H. van Eemeren, J.A. Blair, C.A. Willard, & A.F.  Snoeck Henkemans (Eds.), Proceedings of the Fifth Conference of the International Society for the Study of Argumentation (pp. 641-644). Amsterdam: Sic Sat.
Krabbe, E.C.W., Dalitz, R.J., & Smit, P.A. (Eds.). (1993). Empirical Logic and Public Debate. Amsterdam: Rodopi.
Motley, J.L. (1856). The Rise of the Dutch Republic: A History. New York.
Motley, J.L. (1860). History of the United Netherlands from the Death of William the Silent to the Twelve Years’ Truce, 1609. New York.
Motley, J.L. (1883). The Rise of the Dutch Republic: A History. 3 vols. New York: Harper & Brothers.
Scriven, M. (1976). Reasoning. New York: McGraw-Hill.
Scriven, M. (1987). Probative logic. In F.H. van Eemeren, R. Grootendorst, J.A. Blair & C.A. Willard (Eds.), Argumentation across the Lines of Discipline: Proceedings of the Conference on Argumentation (pp. 7-32). Dordrecht: Foris.
Swart, K.W. (1978). William the Silent and the Revolt of the Netherlands. London: Historical Association.
Toulmin, S. (1958). The Uses of Argument. Cambridge: Cambridge University Press.
Verschueren, J. (1999). Understanding Pragmatics. New York: Oxford University Press.
Wansink, H. (1969). Introduction. In William Prince of Orange 1969, pp. v-viii.
Wedgewood, C.V. (1944). William the Silent: William of Nassau, Prince of Orange, 1533-1584. London: Cassell.
Wellman, C. (1971). Challenge and Response. Carbondale: Southern Illinois University Press.
William, Prince of Orange (1581). The Apologie or Defence of the Most Noble Prince William, by the Grace of God, Prince of Orange. Delft.
William, Prince of Orange (1858). Apologie de Guillaume de Nassau, Prince d’Orange, contre l’Édit de Proscription Publié en 1580 par Philippe II, Roi d’Espagne, avec les Documents à l’Appui. Brussels: Émile Flatau.
William, Prince of Orange (1969). The Apologie of Prince William of Orange against the Proclamation of the King of Spain (H. Wansink, Ed.). Leiden: Brill.
Woods, J. (1992). Public policy and standoffs of force five. E.M. Barth & E.C.W. Krabbe (Eds.), Logic and Political Culture (pp. 97-108). Amsterdam: Royal Netherlands Academy of Arts and Sciences.
Woods, J. (1996). Deep disagreements and public demoralization. In D.V. Gabbay & H.J. Ohlbach (Eds.), Practical Reasoning (pp. 650-662). Berlin: Springer.
Woods, J., & Hudak, B. (1989). By parity of reasoning. Informal Logic, 11, 125-39.

1. Introduction
The concept of conductive argument remains unsettled and controversial in theory of argument. Carl Wellman (1971, p. 52) defined conduction as follows:
Conduction can best be defined as that sort of reasoning in which 1) a reason about some individual case 2) is drawn non-conclusively 3) from one or more premises about the same case 4) without appeal to other cases.

Wellman identified three types of conductive argument: Type One with a single pro reason, Type Two with multiple pro reasons, and Type Three with one or more pro reasons and one or more con reasons. Arguments of the conductive type are clearly non-deductive and, most theorists would argue, non-inductive as well. The term “conductive” indicates a ‘bringing together’ of independent reasons, much like an orchestra conductor brings together many instruments and musicians into a single performance.

The theoretical issues surrounding the concept of conductive argument are almost too numerous to even list in a paper focused on a particular issue. Are all conductive arguments case-based? Should we be talking of conductive evaluations rather than of arguments? Are deductive, inductive, and conductive argument (or evaluation) types an exhaustive and mutually exclusive list? If all conductive arguments are diagrammed as convergent, do we want to say that all convergent arguments are conductive? Even more fundamentally, why should we model various pro and con arguments on a single issue as one conductive argument? There are many other basic questions and issues that could be listed as well.

The focus of the present paper is on the concept of premise weight in Type Three conductive pro and con arguments. Some theorists want to restrict the concept of ‘conductive’ to Type Three pro and con arguments (or evaluations).  The present paper tables that proposal and proceeds on a working hypothesis that understanding the more complex Type Three conductive arguments is a useful pathway for achieving a better understanding of the less complex Types One and Two.

2. Wellman’s ‘Heft’ and Premise Weight
Talk of ‘weighing’ reasons pro and contra is a common manner of speaking. “Premise weight” is an obviously metaphorical expression which some theorists view as an over-stretched and faulty metaphor with respect to its application in theory of argument. For example, Harald Wohlrapp wrote in his Der Begriff des Arguments (2008):
The upshot of the discussion of conductive argument is the following: The conclusion reached with arguments presented is not the result of a weighing, whatever that may be. (p. 333; trans. p. 21)

Trudy Govier is perhaps the only widely known theorist of argument who, in multiple publications, has endorsed and expanded upon Wellman’s concept of premise weight. For Govier, premise weight is not literally measurable, which implies that premise weight must be non-numerical in some sense.

It is important to note that “outweighing” is a metaphorical expression at this point. We cannot literally measure the strength of supporting reasons, the countervailing strength of opposing reasons, and subtract the one factor from the other. (1999, p. 171)

Carl Wellman, the originator of the concept of conductive argument, also seems to have understood premise weight to be non-numerical, as indicated in the following passage from his Challenge and Response (1971):
Nor should we think of the weighing [of reasons] as being done on a balance scale in which one pan is filled with the pros and the other with cons. This suggests too mechanical a process as well as the possibility of everyone reading off the same result in the same way. Rather one should think of weighing in terms of the model of determining the weight of objects by hefting them in one’s hands. This way of thinking about weighing brings out the comparative aspect and the conclusion that one is more than the other without suggesting any automatic procedure that would dispense with individual judgment or any introduction of units of weight. (1971, pp. 57-58)

In this passage, Wellman distinguishes two concepts of weight which might we might conveniently call scale-weight and heft-weight. Scale-weight involves machinery, even if only a simple balance type of scale. The output of the scale-weight process is numerical. Even on a simple balance scale, the use of standard weights can provide numerical weight outcomes. Scale-weight outcomes, being numerical, are precise and absolute rather than non-numerically comparative. Scale-weight is probably the current default meaning of “weight” in both theory of argument and in everyday contexts.

As Wellman, Govier and others have noted, scale-weight is not suitable as the literal basis for the premise weight metaphor. Per Wellman, heft-weight is the correct literal basis for this metaphor, and Govier would likely agree. To my knowledge, heft-weight has not received very much analytical attention in the literature on conductive argument, perhaps because heft-weight is viewed as uselessly vague and subjective.  If this characterization is indeed suitable, then the concept of premise weight in theory of argument falls prey to a destructive dilemma. If scale-weight is the literal basis of the premise weight metaphor, then the metaphor is faulty and over-stretched. If heft-weight is the literal basis of the metaphor, then the metaphor is suitable, but premise weight is thereby uselessly vague and subjective. Perhaps the only way to save the concept of premise weight is to further recharacterize heft-weight. But what would that be like?

In contemplating heft-weight, we can imagine a person lifting several items one at a time and making a verbal pronouncement on each one. Initially the pronouncements will be comparative in nature, such as: much heavier than, heavier than, same weight as, lighter than, or much lighter than. A set of comparative, ranked weight categories is thus progressively created. The objects ranked by comparative weight could then be divided into perhaps five or so categories of non-numerical, verbal weight quantities such as: very heavy, heavy, medium, light, and very light. We need not think of the objects as individually ranked within each weight category, however. The individual human being is here functioning as a comparative weighing machine. Due to the lack of precision of heft-weight, there would be blurred boundaries between categories, and some items would have disputable weight categories, even with just one individual doing the hefting.

The outcome of this individual weighing process is a series of judgments that is objective in the sense that the human body is typically a good, if only approximate, weighing machine that provides a non-numerical, comparative, quantitative output. If one object had a lot more heft than another but a mechanical scale reported the reverse, we would properly believe we had a broken scale. This individual judgment of heft-weight is thus not subjective in the sense of individual personal preferences such as ‘chocolate tastes much better than vanilla’. But is heft-weight valid only for each individual weigher and thus non-objective in the sense of not intersubjective?

It seems to me that heft-weight should be understood as potentially intersubjective and thus objective, despite being non-numerical. As Aristotle noted, the solitary human being is either a beast or a God; so the standard case of Wellman’s ‘hefting’ individual is that he is a member of a group. Let’s say this group has about forty or so people, like the pre-Neolithic human bands, and that there is a mixture of the young and the old, and the frail and the robust. While Wellman’s individual lifter is doing his or her thing, the others are also picking up the same objects in the same way and classifying them into ranked weight categories.

It would soon be found that the mid-range of people in terms of physical ability generally find a group of objects heavy and another group of objects light in weight, approximately speaking. These objects would then become intersubjectively heavy, light, etc. The fact that the Milo’s of this group, the athletically trained weight lifters, found most of the common objects to be light in weight, and the small or frail of the group found most objects to be heavy would all be understood and adjusted for by members of the little group in the usual way. In effect, the mid-range of human strength becomes a kind of standard, much as color words are defined in the standard context of normal daylight. We do not think that red things turn black on a dark night, and we do not think that heavy things literally become light in Milo’s hands.

According to the above account, heft-weight, properly understood is non-numerical, approximate, comparative, and objective (intersubjective). On this characterization, heft-weight has many of the virtues of scale-weight, the major exceptions being lack of numerical output and consequent precision. Instead of numerical output, heft-weight provides non-numerical, comparative quantity categories of an approximate nature. Understood in this way, heft-weight is a very plausible literal basis for the metaphor of premise weight.

It might be objected that approximate, non-numerical quantities are not really quantities at all because quantities are by definition expressed as symbolic numbers. Although such a stance may have numerous defenders, the science of cognitive psychology has recently produced some interesting, and I think relevant, findings about what has been called the approximate number sense. Perhaps the term “quantitative capacity” would have been a better choice here than “number sense”, but the latter wording has taken hold. The distinction between two different quantitative ‘senses’ is more than just a conceptual one. While the symbolic number sense is processed in a spread-out fashion in the prefrontal cortex, the approximate number sense is embodied in another part of the brain called the intraparietal sulcus (Cantlon, et al, 2009) The two number senses seem to be connected in interesting ways. Current research provides preliminary indications that math education can benefit by co-developing the approximate sense and the symbolic number sense. (Halberda et al, 2008) Professional mathematicians are known to exercise their approximate number capacities when socializing at conferences. Classifying the approximate number sense as ‘mere intuition’ is likely an inappropriate over-simplification, given recent findings in cognitive psychology.

A commonly used example of the approximate number sense is when someone views several supermarket lines and classifies them as ‘shortest, short, medium, long, and longest’. Quantities are involved in this process, but typically no counting or symbols. Interestingly, other higher animals have this same ability, which provides obvious evolutionary advantages. The predator needs to choose which group of fleeing herbivores to chase; the fruit-eating animals need to pick which tree will provide the most fruit at the time. It seems quite plausible that this approximate number sense is involved in the process that produces heft-weight. The approximate number sense is comparative, non-numerical, and the product of individual judgment; and heft-weight is all of these things.

Unlike the other higher animals, humans in the process of discriminating quantities obviously verbally characterize the discriminated categories with comparative terms such as ’much more, more, about the same, less, and much less.’ In fact, we do this for a great many types of categories. A very common number of categories in such quantitative verbal hierarchies is three to five to perhaps seven. Seven items apparently are a common maximum quantity for simultaneous cognitive focus in humans. Examples of such additional categories include ‘rich/middle class/poor’, or super rich/rich/upper-middle-class/lower-middle-class/poor’ – and so on. In premise strength, we have ‘strong/moderate/weak’, or perhaps ‘very strong/strong/moderate/weak/very weak’, as categories of discriminated support quantities. Non-numerical quantity categories seem to be essential in human cognition and communication.

In correspondence, Trudy Govier has remarked to me that if the judgment is made to not use “weight” in theory of argument, then “one would have to figure out some other way of speaking. One might speak of deliberating, or comparatively considering, or making judgments of comparative significance.” (1/31/10) I think, and Govier might agree, that these potential substitutions for talk of premise weight would do less work overall than the premise weight concept, understood as heft-weight. We use comparative, non-numerical quantity categories in our reasoning all the time; so dismissing such reasoning as inherently faulty requires a high burden of proof which has not been met.

Non-numerical, comparative quantitative categories are frequently applied by speaking of degrees of this and that. For example, there are degrees of argument strength, degrees of importance, and so on in a great many areas of discourse. In her (2009), Govier has herself puzzled over the so-called ‘degrees’ of argument strength: “What are these degrees anyway? There is no answer.” It seems to me that the principal point of confusion here has to do with “degrees” bringing in symbolic numbers – or not.

Of course, some decision theorists do apply numbers to verbal premise weight categories, e.g. “5” for “very strong”, etc. This approach in my view is best regarded as a ‘game technology’; there are some useful applications for it in contexts of decision making. This ‘invented’ numerical premise weight has no rational basis for conductive argument evaluation for at least one major reason: The exact selection of the number scheme can actually determine the evaluation for some arguments.

To provide just one example, choosing a number scheme of 3-2-1 vs. one of 10-5-2 for the three ‘strong/medium/weak’ verbal categories determines the evaluation of an argument with the following premise weight classifications: four strong pro reasons, five moderate contra reasons, and five weak contra reasons. This type of argument supports its conclusion on a 3-2-1 assignment but not on a 10-5-2 assignment. There is seemingly no way to argue for the rational basis of one number scheme over another for labeling the commonly used verbal categories. Even the total number of quantitative categories is largely contextually determined rather than rule-based. For various reasons, applying numbers to verbal categories has limited theoretical use, if any.

If premise weight determination does not normatively involve the application of symbolic numbers, what positive account of premise weight emerges from the above account? I would argue that premise weight determination involves a classification of each individual premise into one of a small number of non-numerical quantitative categories. With the literal basis of Wellman’s premise weight metaphor, the verbal quantitative categories could be named: ‘very heavy’, ‘somewhat heavy’, ‘medium’, ‘light’ and ‘very light; the corresponding theory of argument categories would be similarly ‘very strong’, ‘somewhat strong’, ‘medium strength’, ‘’somewhat weak’, and ‘very weak.

These non-numerical, quantitative categories of premise weight categories are, to be sure, highly familiar ones. The intent of the above account is to provide them with a clearer grounding than they have previously received, to my knowledge. The fact that the exact names and even total number of such categories is variable and contextually determined is not in my view problematic.

The presumptive weight of an individual premise would in context be based on background knowledge and social values of the individuals and groups involved in argumentation. If a given premise weight is not agreed to, then it can argued for using some version of the scheme for argument to a classification. Premise weights can thus be seen as intersubjectively determinable, contextually and within limits. The contextual reality of deep disagreements is not an effective objection to premise weight as a key term in theory of argument, contrary for instance to Harald Wohlrapp’s critique of Govier on conductive argument.

We shall now apply the above account to some of Govier’s critics on the concept of premise weight and conductive argument, particularly those criticisms focused on quantitative issues. The interpretation of Govier is my own and is of course quite arguable; hopefully it has some measure of accuracy and value.

3. Govier’s ‘Exceptions’ and Issues of Quantification
Govier’s detailed account of weighing reasons is put forward in Chapter 10 of her Philosophy of Argument (1999) and in Chapter 12 of her textbook, A Practical Study of Argument, the current edition being the 7^th (2010). In the first paragraph of her text’s section on conductive argument evaluation, she writes of premises’ “significance or weight for supporting the conclusion.” (p. 359) She soon introduces the specifics of her concept of premise weight, as follows:
While acknowledging that we are dealing here with judgment rather than demonstration, we will suggest a strategy for evaluating reasons put forward in conductive arguments. The premises state reasons put forward as separately relevant to the conclusion, and reasons have an element of generality. That generality provides opportunities for some degree of detachment in assessing the conclusion. Since this is the case, we can reflect on further cases when seeking to evaluate the argument. (2010, p. 361)

Govier’s explication of premise weight uses as its principal example an argument for the legalization of voluntary euthanasia; several of her major critics, including Harald Wohlrapp, have responded to her with further analyses of the same argument, so it is worth stating completely here:
(1) Voluntary euthanasia, in which a terminally ill patient consciously chooses to die, should be made legal.
(2) Responsible adult people should be able to choose whether to live or die.
Also, (3) voluntary euthanasia would save many patients from unbearable pain.
(4) It would cut social costs.
(5) It would save relatives the agony of watching people they die an intolerable and undignified death.
Even though (6) there is some danger of abuse, and
despite the fact that (7) we do not know for certain that a cure for the patient’s disease will not be found,
(1) Voluntary euthanasia should be a legal option for the terminally ill patient.

Govier identifies the associated generalizations for the pro reasons as follows, each with its ceteris paribus clause:
2a. Other things being equal, if a practice consists of chosen actions, it should be legalized.
3a. Other things being equal, if a practice would save people from great pain, it should be legalized.
4a. Other things being equal, if a practice would cut social costs, it should be legalized.
5a. Other things being equal, if a practice would avoid suffering, it should be legalized.

Each generalization is seen to have exceptions, which are the subject matter of the ceteris paribus clause.

For example, you could imagine social practices that would deny medical treatment to medically handicapped children, abolish schools for the blind, or eliminate pension benefits for all citizens over eighty. Such practices would save money, so in that sense they would cut social costs. But few would want to support such actions. Other things are not equal in such cases; the human lives of other people who are aided are regarded as having dignity and value, and the aid
is seen as morally appropriate or required. (2010, p. 361)

The principle of cutting social costs has, in Govier’s terms, a wide range of exceptions.
Perhaps Govier’s most succinct statement about premise strength is in her (1999, p. 171):
A strong reason is one where the range of exceptions is narrow. A weak reason is one where the range of exceptions is large.

For Govier, and within the present paper, the following are treated as roughly synonymous expressions because all are quantitative in a similar way: premise significance, weight, strength, and force. At issue here is the quantitative force of reasons in the broadest sense, as least for Wellmanian ‘type 3’ conductive pros and cons arguments.

Harald Wohlrapp challenges and rejects Govier’s account of a quantifiable range of ceteris paribus exceptions:
But why should the argument be weaker, because the associated if-then sentence has ‘more exceptions’? Can I really compare the number of exceptions through enumeration? Must we not bear in mind that the general principles are situation-abstract and that, depending on how they are being situated, they can have arbitrarily many exceptions? Is there anything countable here? (2008, pp. 323-324; trans. p. 10)

I would like to address this important critique in two respects: (1) issues regarding the nature of these exceptions and in particular their quantifiability; and (2) the general role of the ‘normal situation’ and ceteris paribus in everyday argumentation vs. in scientific contexts. This second issue area will be addressed in Section III of the present paper. What sort of things are these so-called exceptions?

As quoted above, Govier states that the point of framing the generalization associated with a conductive argument consideration is to identify additional cases falling within that generalization. According to Govier, these cases are then to be reflected on in the appropriate process of evaluating premise weight in conductive arguments. Such cases would seemingly be of two kinds, (1) actual cases past or present, and (2) fictional a priori, ‘what if’ cases, including potential future cases. It seems to me that the quantity of exceptions concerns not the number of items on a list of exception categories, which can be almost arbitrarily long. Rather, the quantity of exceptions must involve cases, actual or a priori as described above.

An illuminating question to ask at this point may be as follows: How does Govier come to reasonably believe that there are a great many exceptions to the generalization of cutting social costs? She obviously knows this from her experience living in a wide, but imprecisely delineated, moral community that one might call the developed democracies. She learned about the social values and behavior that create this ‘wide range of exceptions’ by experiencing multiple cases of a normative nature. Two critical questions for Govier’s account are: (1) How and in one sense are such cases counted or numerically assessed, and (2) How and in what sense are such cases relevant to the concerns of normative logic?

Any individual’s knowledge of how many exceptions there are to the principle of reducing social costs is imprecise, which suggests the involvement of the approximate number capacity described above. Explicitly counting exceptions to the principle of reducing social costs is not commonly done. We simply do not go around stating, for example, that there were 794 exceptions to the principle of cutting social costs in the U.S. Congress from 2005 to 2009. Instead, we learn in living which types of cases are very common and which are rare in our moral, legal, and social communities. We do not have in mind the details of most cases and we do not typically count them. We know of a great many cases in which social costs are borne so that other objectives can be attained. We know of comparatively few cases in which unbearable human pain is knowingly tolerated in favor of controlling social costs. Comparative, non-numerical, and individual judgment is being exercised, and that judgment has some objective basis in the quantity of cases comprising the relevant evidence. We acquire knowledge of actual social values by experiencing a great many cases, both legal cases and cases the everyday sense or situations and decisions made. But how are these relevant cases evaluated and processed as evidence, and what concepts and issues within normative logic are involved?

A very fruitful distinction to employ here might be that between case-based legal argument, emphasized in common law-oriented legal cultures, and rule-based legal argument found in civil-law-oriented legal cultures. If I am correct in interpreting Govier’s exceptions-based understanding of conductive argument as a matter of supporting cases in the widest sense of “case”, then the legal model of processing cases, rules and social values may provide insight into the normative aspects of everyday conductive reasoning.

A particularly interesting account of case-based and value-based legal reasoning has been provided Trevor Bench-Capon and George Christie. A legal argument is a paradigm of an argued case. Of course legal arguments and reasoning have been foundational for normative logic since Toulmin. In comparing case-based common law legal argument with rule-based civil law legal argument, George Christie very effectively highlighted the distinctive role of cases in the former:

Under the approach to legal reasoning now to be described [case-based, common law], so-called rules or principles are merely rubrics that serve as the headings for classifying and grouping together the cases that constitute the body of the law in a case-law system. In such a system even statutes are no more than a set of cases, if any, that have construed the statute together with the set of what might be called the paradigm cases that are, in any point in time, believed to express the meaning of the statute. (2000, p. 147)

Arguing from a few precedent cases is of course a standard argument by analogy using the ‘argument from precedent’ scheme. But the picture becomes more complex, and more interesting, once social values are brought in, as theorized by Bench-Capon.

For Bench-Capon, a given case in law is appropriately decided within a key context of often many other cases, past, present and future:
A given case is decided in the context both of relevant past cases, which can supply precedents which will inform the decision, and in the context of future cases to which it will be relevant and possibly act as a precedent. A case is thus supposed to cohere with both past decisions and future decisions. This context is largely lost if we state the question as being whether one bundle of factors is more similar to the factors of a current case than another bundle, as in HYPO, or whether one rule is preferred to another, as in logical reconstructions of such systems. (2000, pp. 73-74)

The context of cases is key because, according to Bench-Capon, “we see a case-based argument as being a complete theory, intended to explain a set of past cases in a way which is helpful in the current case, and intended to be applicable to future cases also. The two goals are closely linked. Values form an important part of our theories and they play a crucial rule in the explanations provided by our theories.”  (2000, p. 74)

Bench-Capon believes that “the ‘meaning’ of a case is often not apparent at the time the decision is made, and is often not fixed in terms of its impact on values and rules. Rather, the interpretation of the case evolves and depends in part on how the case is used in subsequent cases.” (2000, p. 74). Thus case-based argument in law it is commonly not about a small number or cases implying a value scheme but is rather about potentially many relevant cases that modify value schemes in ways not always understood until later interpretations. There is a ‘theory of cases’ that new cases are constantly modifying.

What is the theoretical relevance of these legal arguments, understood as above, to conductive argument evaluation? The factors of legal argument analysis seem to me to be fundamentally the same as the considerations of general pro and con conductive arguments concerned with evaluative issues:
“The picture we see is roughly as follows: factors provide a way of describing cases. A factor can be seen as grounding a defeasible rule. Preferences between factors are expressed in past decisions, which thus indicate priorities between these rules. From these priorities we can adduce certain preferences between values. Thus the body of case law as a whole can be seen as revealing an ordering on values.” (2000, p. 76)

And further:
“In regard to legal theories cases play a role which is similar to the role of observations in scientific theories: they have a positive acceptability value, which they transfer to the theories which succeed in explaining them, or which can include them in their explanatory arguments.” (2000, p. 76)

Cases both express and develop value schemes, which consist of both lists of values and their prioritization in contexts of conflict. Henry Prakken has endorsed this approach as well: “As Bench-Capon [2] observes, many cases are not decided on the basis of already known values and value orderings, but instead the values and their ordering are revealed by the decisions. Thus one of the skills in arguing for a decision in a new case is to provide a convincing explanation for the decisions in the precedents.” (Prakken, 2000, pp. 8-9)

It seems very plausible to me that these points are applicable well beyond legal argumentation. Perhaps weight in conductive arguments, at least those focused on evaluational issues, might best be understood on the model of the above approach to legal case-based arguments. Our daily experience and decisions, both collective and individual, form a kind of case history which both expresses and continually forms and re-forms our values. Philosophers in recent decades have tended to understand moral issues (and sometimes practical issues) in terms of rule-based models rather than in terms of case-based models, but this long-term emphasis may have been overdone. It seems to me quite plausible that the case-based reasoning model would readily apply to non-moral, evaluative, conductive reasoning as well.

The idea of value schemes evolving with case decisions is entirely consonant with Stephen Toulmin’s remarks in The Abuse of Casuistry: “Historically the moral understanding of peoples grows out of reflections on practical experience very like those that shape common law. Our present readings of past moral issues help us to resolve conflicts and ambiguities today”. (1988, p. 316) It seems to me that taking the case-based understanding of legal reasoning, together with modeling much everyday evaluative reasoning on legal argument interpreted as value-centric, is a very promising direction.

Perhaps a very broad characterization of the type of reasoning in question might be what Robert C. Pinto and others have called “support by logical analogy”. In his (2001, p. 123), Robert C. Pinto describes the method of logical analogy as “pre-eminently important.”  Pinto further notes: “Though it [argument from logical analogy] is fairly widely recognized as a method for justifying negative evaluation of arguments and inference, in my view it can also provide grounds for positive evaluations as well.” Govier addresses refutation by logical analogy in her textbook’s chapter on analogical reasoning. I am not aware of her addressing support by logical analogy elsewhere. David Hitchcock has written a very interesting paper (1994) on conductive argument validity which utilizes, according to my understanding of it, refutation by logical analogy; I believe he does not address “premise weight” here specifically. The point I would like to add is that support by logical analogy would seemingly involve analogous cases that might be argumentatively addressed in the mass, rather than in the substantial detail of a standard two-case argument by analogy.

It might be objected that in focusing on Govier’s talk of further cases to reflect on, I am hopelessly blurring the distinction between conductive and analogical argument. The claim that premise weight is commonly supported by, broadly speaking, analogical types of arguments does not imply that conductive arguments are types of analogical arguments. The main argument, the first tier of reasons above the conclusion (the main conclusion being at the bottom of the argument diagram), may be convergent but have analogical subarguments either in the dialectical tier or in corresponding evaluation arguments. It is interesting to note that analogical and conductive arguments are typologically ‘cousins’ in a sense in that both are inherently comparative in nature.

Not all conductive arguments are about valuational matters. Some theorists’ efforts regarding the ‘quantity of evidence’ in conductive argument might best be seen as regarding conductive arguments with non-valuational conclusions rather than conductive arguments in general. For instance, in his Cognitive Carpentry, John L. Pollock proposed numerical quantitative assignments to premises for arguments that can be interpreted as statistical syllogisms. In his (2002), Alexander V. Tyaglo has applied probability theory to separate reasons in convergent arguments. The epistemic status of the probability numbers themselves makes this approach one of limited scope and value.

Ideas from Pollock and from Tyaglo may be applicable to predictive (or dispositional) conductive arguments that seem to be arguments from sign. An example of such an argument appears early in Govier’s textbook chapter on conductive argument: “She must be angry with John because she persistently refuses to talk to him and she goes out of her way to avoid him. Even though she used to be his best friend, and even though she still spends a lot of time with his mother, I think she is really annoyed with him right now.” (2010, p. 366) Whether it is useful to identify two (or more?), subtypes of conductive argument, the empirical and the valuational, is an interesting question worth pursuing. The argument of the present paper concerns principally ‘valuational’ conductive arguments.

4. Cumulating Independent Reason Strands
The above account characterizes premise weight determination as normatively involving a scheme of argument to classification among a small number of non-numerical but quantitatively ranked categories, i.e. ‘very strong’, ‘strong’, etc. This claim is of course not at all novel. The present intent is to provide additional conceptual support and clarity for the concept of degrees of premise weight and argument strength. What is excluded for those who accept the above account is the view that premise weight is either entirely subjective or entirely objective, as would be implied by accepting the scale-weight model of premise weight or by rejecting the concept of premise weight altogether. The above account thus supports a middle ground of intersubjectivity.

Most of the above account has to do with the concept of individual premise weights. But, how are the various reason strands of a given argument to be normatively ‘conducted’ together into an evaluation of their net collective support, or lack thereof, for an argument’s stated conclusion? More ‘dustbin empiricism” might be helpful here in order to better develop what Robert C. Pinto calls critical practice, an aspect of which would here be a checklist of questions as a guideline to good conductive argument evaluation.

It seems to me that, descriptively, people commonly begin a conductive argument evaluation by viewing the whole argument and classifying considerations as major or minor. Ben Franklin famously crossed out opposing, equally (heft-) weighted considerations. Descriptively, it seems to me that we seem to hold those considerations identified as “minor” in reserve, in case there is a perceived ‘tie’ between the major considerations on each side. Arguments with, for instance, two strong pro premises, one weak pro premise, and two strong con premises may just be unresolvable, unless more considerations can be added or individual premise evaluation differences resolved by the arguers. But such common-sense observations and guidelines hardly constitute an example of adequate theory of argument.

It may very well turn out that normative logic has rather little to offer in terms of addressing premise cumulation in conductive argument. Harald Wohlrapp famously argues exactly this point and offers his dialectical frame-integration account of resolution. But it seems to me that his approach rings true because it brings in values; a frame for Wohlrapp is a valuational perspective on a set of characterized (or recharacterized) facts. Addressing values directly is, as previously mentioned, also a feature of legal case-based, value-based reasoning. Values are commonly brought into contexts of everyday conductive argument as well.

5. Conclusion
A longer paper would have been able to further address a number of issues regarding premise weight. For example, the concept of ceteris paribus and the ‘normal situation’ highlighted in Govier’s account deserves more extensive treatment. Also deserving of attention is Frank Zenker’s interesting proposal that (1) deductive, inductive and conductive arguments all have premise weights, but that (2) the premise weights in deductive and inductive arguments are ‘equal’ and thus in a sense tacit. (Zenker, 2010) Perhaps the concept of premise weight could be useful in clarifying evaluation typologies along the following lines: (a) deductive evaluation is structural with equal-weight reasons; (b) inductive evaluation is additive (or cumulative) with equal-weight reasons; and (c) conductive evaluation is comparative with, unequal-weight reasons.

Overall, the logic of conductive argument remains somewhat obscure, but perhaps we are collectively making some small progress. A main take-away from the present paper, in my view, is that the concept of premise weight is a fruitful one that is entirely worthy of contemporary interest and further investigation in theory of argument.

REFERENCES
Bench-Capon, T., & Sartor, G. (2000). Using values and theories to resolve disagreement in law. In J. Breuker, R. Leenes & R. Windels (Eds.), Legal Knowledge and Information Systems. Jurix 2000: The Thirteenth Annual Conference (pp. 73-84). Amsterdam: IOS Press.
Bench-Capon, T. (2001) Review of: The Notion of an Ideal Audience in Legal Argument. Artificial Intelligence and Law, 9(1), 59-71.
Cantlon, J., Libertus, M., Pinel, P., Dehaene, S., Brannon, E., & Pelphrey, K. (2009). The neural development of an abstract concept of number. Journal of Cognitive Neuroscience, 21(11), 2217-2229.
Christie, G. (2000) The Notion of an Ideal Audience in Legal Argument. Dordrecht: Kluwer Academic.
Govier, T. (1999). The Philosophy of Argument. Newport News, VA: Vale Press.
Govier, T. (2009). More on dichotomization: flip-flops of two mistakes. In J. Ritola, Ed., Argument Cultures: Proceedings of OSSA Conference 2009. CD-ROM, Windsor, ON. June 2009.
Govier, T. (2010). A Practical Study of Argument, 7^th edition. Belmont, CA: Wadsworth Cengage Learning.
Halberda, J., Mazzocco, M., & Feigenson, L. (2008). Individual differences in non-verbal number acuity correlate with maths achievement. Nature 455, 2 October 2008, 665-668.
Hitchcock, D. (1994). Validity in conductive arguments. In R. Johnson & J. A. Blair   (Eds.), New Essays in Informal Logic (Ch. 5, pp. 58-67). Newport News, VA: Vale Press.
Jonsen, A., & Toulmin, S. (1988). The Abuse of Casuistry. Berkeley: University of California Press.
Kock, C. (2007). Is practical reasoning presumptive? Informal Logic, 27, 91-108.
Libertus, M., & Brannon, E. (2009). Behavioral and neural basis of number sense in infancy. Current Directions in Psychological Science 2009, 18(6), 346-351.
Pinto, R. (2001). Cognitive science and the future of rational criticism. In Argument, Inference, and Dialectic (Ch. 12, pp. 113-125). Dordrecht: Kluwer.
Pinto, R. (2009). Argumentation and the force of reasons. In J. Ritola (Ed.),Argument Cultures: Proceedings of OSSA Conference 2009. CD-ROM, Windsor, ON. June 2009.
Pollock, J. (1995). Cognitive Carpentry. Cambridge, MA: MIT Press.
Prakken, H. (2000) An exercise in formalizing teleological case-based reasoning. In J. Breuker, R. Leenes & R. Windels (Eds.), Legal Knowledge and Information   Systems. Jurix 2000: The Thirteenth Annual Conference (pp. 49-57). Amsterdam: Ios Press.
Tyaglo, A. (2002). How to improve the convergent argument calculation. Informal Logic, 22(1], 61-71.
Wellman, C. (1971). Challenge and Response. Carbondale, IL: Southern Illinois University Press.
Wohlrapp, H. (1998). A new light on non-deductive argumentation. Argumentation, 12(3), 341-350.
Wohlrapp, H. (2008).  The pro- and contra-discussion (A critique of Govier’s “conductive argument”). Translation by Frank Zenker of: Wohlrapp, H. (2008) Der Begriff des Arguments, Section 6.4. Wurzburg: Konigshausen und Neumann. Retrieved from www.frankzenker.de/academia.
Zenker, F. (2007). Complexity without insight: Ceteris paribus clauses in assessing    conductive argumentation. In S. Jacobs (Ed.), Proceedings of the 2007 NCA/AFA Conference on Argumentation, Alta, Utah, 810-181.
Zenker, F.  (2009). Ceteris paribus in conservative belief revision. On the role of minimal change in rational Theory Development (Ph.D. Thesis, University of Hamburg). Berlin: Peter Lang (ISBN 978-3-631-57283-2).
Zenker, F. (2010). Deduction, induction, conduction: An attempt at unifying natural     language argument structure. In Proceedings of the 13^th Biennial Argumentation Conference at Wake Forest University, March 2010.
Zimmer, C. (2009). The math instinct. Discover Magazine. November, 2009, p. 28.

1. A Case of Unreconciled Dissensus
Book V of Milton’s Paradise Lost presents a striking dissensus between Satan and the Archangel Abdiel over the nature of the Deity. Each presents an argument for his view which – not unsurprisingly – the other rejects. Milton sets the scene – The Almighty before a convocation of all angels has decreed his Son their Lord and has mandated that “to him shall bow/All knees in Heav’n, and shall confess him Lord” (V, 607-608) This decree Satan cannot abide. He resolves to rebel, never bending the knee, nor, if he can persuade them, will any of the angels under his command. Paraphrasing to bring out the underlying argument, Satan first proposes
(1) Prior to this decree, all Natives of Heaven (including the Almighty and his Son) have been equally free.
(2) No one has a right to assume monarchy over one’s equals in freedom. Hence
(3) The Almighty has no right to proclaim this decree.

Although Satan offers two further arguments, Abdiel turns his critical questions exclusively to Satan’s first. Again paraphrasing, his argument can be laid out quite straightforwardly:
(1) The Almighty created you and indeed all the spirits of heaven, and endowed all with their glory. Therefore
(2) Neither you nor all angels taken together are equal to the Almighty. Therefore
(3) Justice gives you  no right to enter with God in determining what are the laws or principles governing your relation. Therefore
(4) The Decree of the Almighty is just.

Satan replies first by questioning Abdiel’s first premise. What evidence is there for this creation, he asks. Who observed it? Do your remember your own making? Satan then continues
We know no time when we were not as now;
Know none before us, self-begot, self-rais’d
By our own quickening power….
(V, 859-861)

These observations bear on his assertion that “Our puissance is our own,” i.e. we are not creatures of or subordinate or inferior to the Almighty. Satan ends his discourse by ordering Abdiel quickly to report his sentiments to the King. The dialectic thus ends at this confrontation stage.

With passions running as high as Milton portrays them, one wonders whether the argument could be advanced to a further stage. However, even assuming dispassionate interlocutors, the literary critic and legal scholar Stanley Fish has argued that it could never proceed to a rational resolution. Since his argument presents a challenge to the whole enterprise of argumentation, it deserves the attention of argumentation theorists.

2. Fish’s Challenge to Argumentation
In arguing that rational resolution of their dispute is impossible, Fish focuses on Satan’s asking Abdiel to show that we are created beings and construes the passage, already quoted,
We know no time when we were not as now;
Know none before us, self-begot, self-rais’d

as an argument, our self-creation being inferred from our lack of knowledge of a time when we were other than as now. Fish asks us to contrast this argument with that of the newly created Adam, aware for the first time both of his surrounding world and its beauty and of his body with its powers:
But who I was, or where, or from what cause
Knew not, …
… how came I thus, how here?
Not of myself; by some great Maker then,
In goodness and in power preëminent;
(VIII, 270-271, 277-279)

Fish sees Adam arguing from the premise that he does not know how he came into being to the conclusion that he owes his being to a Maker first in goodness and power. In the context of his argument that all the angels are creatures of the Almighty, Abdiel has made a remark whose relevance he might have highlighted should Satan have permitted him to give evidence of that claim:
Yet by experience taught we know how good,
And of our good, and of our dignity
How provident he is, …
(V, 826-828)

Adam and Abdiel’s reasoning share this epistemological point:  Our inferences may pass beyond the realm of experience in finding an explanation of the experienced realm or seeing some significance, e,g, the Deity’s benevolent nature, which it points to.  By contrast, Satan rejects  both inferences a priori.

Fish sees both arguments as incompletely stated, both lacking a first premise. Given recent work on enthymemes,[i] I believe it better to say that both arguments instance substantial, as opposed to formal, inference rules or warrants.

Satan’s warrant:
Given that x is consciously aware of no time when x was other than as now nor of any predecessor or progenitor of x
One may take it that x is self-created

Adam’s warrant:
Given that x knows not how x got to this place of preëminent beauty possessed of a body of preëminent vitality
One may take it that x is the work of a Maker unsurpassed in goodness and power.

Fish now makes a crucial point for his argument that this exchange between Satan and Abdiel cannot go beyond the confrontation stage:
Since the first premise is what is missing, it cannot be derived from anything in the visible scene; it is what must be imported – on no evidentiary basis whatsoever – so
that the visible scene, the things of this world, can acquire the meaning and significance they will now have. (Fish 1996, p. 19, italics in original)

It is a commonplace that corresponding to an argument is a conditional statement, the conjunction of the premises being the antecedent, the conclusion the consequent. As Hitchcock (1985) has shown, arguments which some analyze as first-order enthymemes assume more than this associated conditional, namely some universal generalization of that conditional. As we have argued (2011), this universal generalization must be nomic, supporting subjunctive conditionals, and not merely accidental. It is never a description, an extensional statement whose truth conditions concern just the actual world. In many instances, it is an interpretation,[ii] an intensional statement whose truth-conditions involve considering other possible worlds.[iii] Hence, if to be derived from the visible scene means simply to describe some aspect of one’s surroundings of which one is aware just through sense perception, we agree with Fish that the first premise cannot be derived in this way. We also agree that in the light of interpretive generalizations, certain descriptive features acquire meaning (or their meaning becomes disclosed). This point may be appreciated better in connection with warrants. Consider again Adam’s warrant. Although the premise involves an aesthetic evaluation rather than a mere description, in light of this warrant Adam does not see himself in a randomly beautiful world but in one whose beauty is attributable to conscious agency. But if one has an explanation for some event or condition, that event or condition has meaning, at least in some sense or to some degree. Likewise, Satan’s warrant is interpretive. It associates a meaning, being self-created, with the non-awareness of one’s origination or of any originating progenitor.

Fish elaborates his position that first premises – alternatively warrants – cannot be based on evidence by saying
In the absence of a fixed commitment–of a first premise that cannot be the object of thought because it is the enabling condition of thought–cognitive activity cannot get started. One’s consciousness must be grounded in an originary act of faith – a stipulation of basic value – from which determinations of  right and wrong, relevant and irrelevant, real and unreal, will then follow. (Fish 1996, pp. 19-20)

Following Fish, let us refer to this as the Miltonian position. Hence we understand the position asserting that by virtue of our warrants, we recognize what is relevant to what, that something’s possessing a certain property is evidence that it possesses some further property, but that these warrants as principles of evidence are not themselves defendable through evidence and thus not defendable through argument. They are and must be accepted on faith, the faith constituting at least part of  one’s world view. One might say that warrants used in particular arguments derive in some sense from some fundamental warrant or warrants. But those basic warrants are not based on any evidence, their acceptance being an act of faith.

Continuing within the framework of the Toulmin model, we see another point at the core of the Miltonian position. Recall that non-demonstrative warrants are open to rebuttal. We have already seen that it is part of Satan’s epistemological stance to recognize as real only what is disclosed by descriptive belief-generating mechanisms analogous to perception, memory, introspection. Hence, any warrant permitting us to infer something non-observable from what is observable must be rejected. The principle identifying “experience” with being is a blanket rebuttal of all such warrants. Again, such a rebuttal cannot be defended with evidence, but derives from the basic act of faith which stipulates what is real and unreal. Warrants, then, as constituting principles of evidence, and rebuttals, as ruling out certain inferential moves, are articles of faith, not subject to critical scrutiny or support through argumentation.

Fish sees in this picture of the structure of cognitive activity a challenge to the liberal ideal of open mindedness to all positions, including those incompatible with one’s cherished opinions, an open mindedness including a willingness to revise one’s viewpoints in light of argumentation. As such, the picture challenges much of the argumentation community’s understanding of the practice of argument and its ideal conditions. For example, consider the pragma-dialectical code of conduct for rational discussants. Van Eemeren and Grootendorst require that “the discussants must be able to advance every point of view and must be able to cast doubt on every point of view” (1984, p. 154). If asked, a party advancing a standpoint must defend it with cogent argument. If the defense fails, the proponent must retract the standpoint. If it succeeds, the challenger must retract her doubt. (Compare Rules 2 and 9 in (1992, pp. 208-209).) Clearly, on Fish’s picture if one tried to argue for a claim expressing the propositional content of a warrant one accepts, one would at best be arguing in a circle. Since the warrant determines what is deemed relevant or irrelevant, the very warrants one’s argument would instantiate would ultimately be acts of faith. Any proponent who realizes this realizes that he cannot argue cogently for that claim.[iv]

Even if the proponent failed to realize the futility of his attempted argument, it is hard to see how the discussion could ever proceed to the argumentation stage. This stage presupposes agreement on the rules of discussion. But if proponent and challenger have different, indeed incompatible originating acts of faith concerning their warrants, their very inference rules and rebuttals, grounded in such originating acts of faith, will differ and essentially differ. Remember these originating acts of faith are not subject to rational appraisal. Even if the parties attempted to bypass agreement on rules and proceed to argumentation, I do not see how the proponent could realize that his argument failed, if it did,  or the challenger realize that the proponent’s argument was successful, if it was. If the proponent’s argument depends on an inference rule the challenger does not accept or the proponent would not recognize the force of the challenger’s rebuttal, the discussion could never reach the concluding stage. A critical discussion in the pragma-dialectical sense is impossible on the Miltonian position.

For the Miltonian, the belief expressing the faith of the originating act constitutes what is understood as reasonable by the person making that act of faith. Any viewpoint challenging that originating belief will be dismissed as unreasonable. “A reasonable mind is a mind that refuses to be open” (Fish, 1996, p. 20). Fish sees this Miltonian stance as typifying religious commitment, the shared faith of a religious community. Indeed, we might see it as typifying ideological commitments in general, and more generally as typifying world-view commitments. For the adherents of a religious tradition or an ideology with a core creed, challenges to the tenets of that creed might seem impossible. Again, a challenge to any facet of one’s world-view would seem absurd.

The liberal stance presupposed by argumentation theory’s very understanding of argument as dialectical seems incompatible with the Miltonian stance of commitment. To seek to resolve a difference of opinion through argument, the parties must agree on the principles of evidence certifying the outcome. But especially if the difference concerns some opinion central to the world- view of one of the parties to the discussion, and world-views determine the acts of faith which determine principles of evidence, a dialectical discussion seems impossible. But to what extent are differences of opinion the result of differences over principles of evidence? Perhaps not all differences of opinion involve such differences, and this leaves a door open for the liberal view of argument.

One way for the advocates of argument to deal with this dissensus over world-view commitments would be to rule out argument over those commitments or over opinions essentially deriving from them, and to rule out appealing to any principles of evidence essentially dependent on them in any dialectical exchange, at least in any dialectical exchange in the public sphere. Not only does this accord with a liberal stance, Fish argues that it itself actually expresses a core ideological commitment of liberalism:

Liberalism rests on the substantive judgment that the public sphere must be insulated from viewpoints that owe their allegiance not to its procedure – to the unfettered operation of the market-place of ideas – but to the truths they work to establish. (Fish 1996, p. 22)

Liberalism presupposes that at least some issues of fact and principles of evidence can be disentangled from issues of ideology. That “a stage of perception…exists before interest kicks in” is a “prime tenet of liberal thought” (Fish 1996, p. 25). For liberalism, we might say, a viewpoint not justifiable through principles independent of ideological commitments cannot be taken seriously. It is as unreasonable from the liberal point of view as the viewpoints challenging that view are unreasonable from the viewpoint of those committed to that viewpoint.

If this characterization of liberalism is correct and the argumentation community is committed to the liberal stance, then it would seem that the argumentation community is intolerant of ideological commitment, including religious commitment. Such commitments are beyond the pale of argumentation and attempts to resolve them through argument futile. Such a viewpoint may well have negative social consequences for the argumentation community. It suggests that most of the commitments by which persons see meaning and value fail to be rationally grounded, with all the negative emotive force of that characterization. Those with world-view commitments who might take umbrage over this characterization have a riposte. Liberalism’s commitment to principles of evidence regarded as independent of world-view commitments and rejection of ideologically dependent principles is simply part of its ideological commitment! Liberalism is an ideology on all fours with other ideologies, but involving this distinct paradox: Liberalism’s core principles concerning evidence are originating ideological commitments not subject to justification through evidence and therefore contradictory to those very principles themselves! How may we come to the rescue of argumentation?

3. Is Argumentation Caught in a Dilemma?
Let us say first that Fish’s epistemological view contains a very important insight, one which I believe he shares with Peirce. (See “What is a Leading Principle” in (1955), pp. 129-134.) Peirce analyzes belief as a habit which develops under the stimulation of various experiences and the pathways we find most successful in dealing with these irritations. One type of belief-habit conveys us from one judgment, the premise, to another judgment, the conclusion, i.e. the belief-habit allows us to infer the conclusion from the premise. Clearly, since the experiences of different individuals will be different, we may expect them to develop different habits, including different inferential belief-habits. These differences will affect intuitions of what counts as a reason for what, intuitions of relevance. Hence we find Fish on solid ground when he allows that different persons will recognize evidence differently. To be able to infer a conclusion from a premise is to recognize that the premise or what it expresses has a certain meaning. Different persons then will recognize meaning differently and interpret situations differently. But we cannot agree that the first premise of any argument  is imported or must be imported “on no evidentiary basis whatsoever.” Taking the assumption as a warrant rather than a premise, Fish in effect is claiming that no warrants can be backed, in Toulmin’s sense, more generally that they and their associated nomic universal generalizations are immune to logical or epistemological evaluation. Is this true? Are they simply matters of faith?

By including backing for warrants in the layout of arguments, Toulmin is allowing that warrants are subject to evidentiary support. As is well known, given his notion of argument fields, Toulmin allows distinctly different types of such evidentiary support.[v] But this does not gainsay the fact that warrants can be supported with evidence. Indeed the very considerations showing that Peirce and Fish would agree that different persons reason according to warrants belonging to different classes also shows that they would disagree on warrants not having evidentiary support. The experiences which led to the formation of the belief-habit constitute evidentiary backing for it. Furthermore, as Toulmin has taught us, not only can warrants be backed, they can be rebutted. But this is to bring negative evidentiary considerations to bear on evaluating the reliability of the warrant. Further yet, a challenger may raise the question of whether a rebuttal holds and a proponent may show that it does not, thus giving a further type of evidentiary support to the warrant.

Pace Fish, we can subject both Satan’s and Adam’s warrants to rational scrutiny. Consider the premise of Satan’s warrant:
x is consciously aware of no time when x was other than as now nor of any predecessor or progenitor of x.

Substituting for ‘x’ a referring expression denoting some being with a capacity for memory, the intended domain of this warrant, produces a logically consistent statement. There is nothing self-contradictory in saying
John is consciously aware of no time when John was other than as now nor of any predecessor or progenitor of John.

But consider the conclusion–John created John. Is the notion of a self-created being logically consistent?  Although this, like all substantive philosophical positions, is open to debate, common sense might vote that self-creation is not coherent. But surely a warrant allowing one to pass from a consistent statement to one metaphysically incoherent is totally unreliable, if not invalid. That no being can create itself constitutes a serious rebuttal to Satan’s warrant. By contrast, Adam’s warrant is abductive, passing from a description/evaluation to an explanation. But one can certainly argue for an explanation by arguing that it is superior to its alternatives, which constitute possible or potential rebuttals.  Such an argument, better the evidence included in the premises of the argument, constitute evidence for the warrant. Although Adam may reason according to his warrant without reflection, this in itself does not show that his warrant can only be accepted on faith.

Fish may now object that the critique betrays a superficial understanding of his position. Satan’s warrant derives from his “faith” that the limits of his experience determine the limits of reality This faith is essential to Satan. “The habit of identifying the limits of reality with the limits of his own horizons defines Satan – it makes him what he is” (1996, 19). Since you do not share Satan’s essential commitment, you may judge that Satan’s warrant may be rebutted. But you yourself have essential commitments, or at least commitments to one or more overarching basic or first principles, not open to your consideration because they determine the very structure of your rationality, including your capacity to critique other viewpoints. Fish endorses this position in a striking epistemological statement:
Evidence is never independent in the sense of being immediately perspicuous; evidence comes into view (or doesn’t) in the light of some first premise or “essential axiom” that cannot itself be put to the test because the protocols of testing are established by its pre-assumed authority. (1996, 23)

Is this true? Suppose one’s experience leads to forming an inferential belief-habit expressible as a warrant. Suppose one meets another whose stock of inference habits does not include this warrant. If one presents the evidence or paradigm instances of the evidence which led to the forming of one’s belief habit, why cannot the other appreciate that they constitute positive evident for that warrant, and indeed may even constitute sufficient evidence for acceptance? How is some essential axiom necessary to recognize this evidence as evidence? Again, on what essential axiom does one’s recognition of the incoherence of a self-created being rely? The newly created Adam could have entertained an additional hypothesis in considering how he came to be in the environment in which he found himself with his body having the powers he is aware of. It all just popped into existence by chance. Does Adam need an originating faith to see which hypothesis he is aware of has higher probability? What essential axiom is necessary for him to see that given two rival hypotheses, the one with the greater likelihood is the one better supported by the evidence–the prime principle of confirmation?

Let us return to the confrontation between Satan and Abdiel. Satan believes he is the equal of the Almighty, at least in freedom. Abdiel believes he is a creature of the Almighty, and thus not equal. These “articles of faith” have a bearing on why Satan accepts the warrant
Given that x has declared the son of x Lord over all Y’s
One may take it that x has made a power grab

while Abdiel does not. Satan and Abdiel thus differ radically on the meaning of the event and thus on whether their experience constitutes evidence for their contrary interpretations. Now there is a profound epistemic difference between saying that the Deity made a certain proclamation and saying that by making this proclamation the Deity made a power grab. The first is a simple description of a publically observable event. The second is a claim about the intentions of the Deity, not open to public inspection. That Satan’s and Abdiel’s different views on the intentions of the Deity are due to fundamental differences in their originating commitments over their creaturely status constitutes a plausible explanation for their dissensus. By virtue of their different originating commitments, they interpret experienced features of reality differently. Could one amend the Miltonian claim to allow that accepting principles of evidence for descriptions of observable events may be independent of any originating commitment, together with recognizing when broadly logical concepts hold and making judgments or estimations of probability, but that accepting principles of evidence involving interpretive principles, including evidence for those principles themselves, is consequent upon an originating commitment?

Such an amendment constitutes a significant concession for the Miltonian to liberalism. Some principles of evidence may be disentangled from ideology. But if our examples of experiential backing for warrants, considerations of the incoherence of self-causation, or best explanations for evidence are cogent, we do have some sources of objective evidence and  objective critique of principles of evidence. Hence, although we can agree with Fish that many rules of evidence one person acknowledges may differ from the rules of evidence acknowledged by someone else, and we can also agree that a person’s commitments, especially in connection with value, ideology, and world view, issue in a set of inference habits specifically reflecting those commitments, we do not agree that these need to constitute the entire set of evidence principles and  inference habits a person employs.

However, excluding argumentation from a significant role in the areas of meaning and value may make its role and the liberalism it expresses seem quite thin. Do most arguments in the public sphere confine themselves just to descriptions and the generalizations they support,  assertions about broadly logical relations, or estimations of probability and their epistemic consequences? Do not the balance of arguments in the public sphere concern meaning and value? The Miltonian can urge: True, you have shown that there are principles of evidence independent of originating commitments. But by contrast with the big existential questions, are not the issues of these arguments superficial? Contrast such concerns with the commitments of Satan and Abdiel. For Satan, the world, as disclosed to us by our experience, is all there is, and this experience, in itself, discloses no being on whom the world is metaphysically dependent. This core commitment determines his refusal to acknowledge any creaturely dependence. Hence any worship of another is“prostration vile” (V, 782). By contrast, at the core of Abdiel’s world view is acknowledgment of creaturely dependence on the Almighty and trust in his providence. Are not these contrasting world views each the product of radially different originating commitments? But if you concede that argumentation cannot deal with dissensus over such world-view issues, you have made a great concession to my Miltonian position.

But why are Satan’s and Abdiel’s contrasting metaphysical beliefs  immune to scrutiny on the basis of commonly recognized epistemic principles of evidence?  Do ideological or metaphysical commitments and what they entail always lie outside what can be subject to critical discussion? Can argumentation play no role in adjudicating such disagreements? We turn to that issue in the next section.

4. Can Argumentation Not Deal With Certain Cases of Dissensus?
As Fish has indicated, these metaphysical commitments constitute “an originary act of faith” from which judgments of meaning and value follow. The propositional content of such an act of faith is some ultimate premise or “essential axiom.” The warrants we apply in the “lower level” arguments we have been considering or the associated universal generalizations of these warrants are consequences of these essential axioms. It is by virtue of subscribing to some essential axiom that we recognize some statement as evidence for some other. In addition to the examples of evidentiary relations we have been considering – particular instances supporting and thus backing generalizations, recognition of broadly logical entailment and related concepts such as coherence or incoherence, recognition of relations of conditional probability – we may add recognition that certain descriptive properties such as having made a promise are relevant to certain evaluative properties, here being morally bound to fulfill it.

As we have seen, our previous considerations here cast real doubt on Fish’s claim that recognizing relevance, i.e. recognizing what constitutes evidence for what, is dependent on originating commitments. We can raise the same issue for Fish over lower level arguments of value. How are originating commitments involved in seeing that my making a promise is a reason why I am bound to keep it, at least a prima facie reason from which my obligation follows ceteris paribus? If someone disagreed about the obligation or just failed to see it, one might invite the person to carry out a thought experiment, imaginatively entering into a situation with the same deontically relevant properties, where that person would admit that the obligation was binding. But where does some essential axiom enter into this argument? The burden of proof, we may urge, is on Fish to show in all these lower-level cases how the recognition of evidential relevance derives from some essential axiom and would be impossible without the recognition of such an axiom. In light of the fact that expecting agreement over relevance in many lower-level cases seems straightforward, Fish has a heavy burden of proof. We shall see the import of this point shortly.

One strategy Fish might use to discharge this burden of proof would be to argue that we are being provincial. We are simply assuming that our recognitions of evidentiary relevance are universal. The fact that we can confidently expect agreement on judgments of relevance only shows that we have confined our circle of acquaintance to those sharing our originating act of faith or some basic principle overlapping with it significantly. That explains our intuitions of relevance and expected consensus. But imagine someone who holds that our making a promise is not much of a reason for saying we are obligated to keep it. Indeed, suppose the person held that our perceiving where  making a promise with no intention to keep it would advance our self-interest in a given situation, we have reason to do just that. Now we are faced with someone with a different essential axiom from which it does not follow that making a promise is relevant to keeping it, or that self-interest always trumps moral regard for others. How would you argue with that person?

This question gains significant poignancy in light of our diverse world. People do disagree on fundamental commitments–for or against democracy as the proper form of government, for or against seeing the human individual as having a value superior to the human collective, for or against seeing facts in the world having a transcendental import. Can argumentation deal with dissensus over such commitments, which we may call world-view commitments? It is here that our considerations on recognizing evidentiary relations independently of world-view commitments come to the fore. We may see world view commitments providing an overall, overarching, or comprehensive explanation, investing events in the world with meaning, or setting limits on the scope of any explanation. We have already seen how Satan’s view of reality as co-extensive with experience and of himself and his angels as self-made led to radically different value commitments from Abdiel’s view of his creaturely status. Given conscious recognition of a world-view, then, one is confronted with two sources for one’s judgments of evidentiary relevance – one’s individual recognition of relevance apart from any world – view commitment and judgments deriving from that commitment. Where such judgments agree, they are mutually reinforcing. Where they do not, adjustment either on the part of the world-view commitment or on the part of certain individual judgments or both is required to maintain consistency. The goal is to reach what Rawls calls reflective equilibrium. The point is that when in reflective equilibrium, there is a mutually reinforcing evidentiary relation between the world-view commitment and the individual judgments of relevance. “From below,” the individual judgments support the “essential axiom” of the world-view commitment. “From above,” that the individual judgments may derive from such an axiom supports such judgments. World-view commitments may then be supported by evidence and it seems we may recognize these support relations independently of the commitment.

We may now address the question of what should be the function of argumentation when dealing with world-view dissensus. Clearly, although complete reflective equilibrium may be an ideal, we expect that in actual cases equilibrium will be a matter of more or less. The more equilibrium, the greater the evidential support, the less the lower. Clearly also, ceteris paribus, reflective equilibrium is a sign of the reasonableness of both the fundamental commitment and the individual judgments, and a system in which there is greater reflective equilibrium is one with greater reasonableness. When persons or cultures with divergent world-views meet, they may be able then, to recognize the reasonableness of each other’s world view commitments through recognizing degree of reflective equilibrium. An argument which prima facie showed why one’s world view commitments functioned as basic principles for one’s judgments of meaning and value would be a case for the prima facie reasonableness of both the world view commitments and the judgments of meaning and value. Surely such an argument could be appreciated as prima facie reasonable by someone not sharing those commitments, and indeed such an appreciation would be an act of respect and deepening respect for those who do hold these commitments. But here is an obvious role for argumentation.

The role of argumentation goes further. Those holding one world view might come to recognize that the basic commitment, essential axiom of those in some other culture may possibly be in better reflective equilibrium or hold promise of better reflective equilibrium with their own individual judgments than their own basic axiom. Greater reflective equilibrium would be possible by either accepting the other culture’s basic axiom or by modifying their own essential axiom to approximate that of the other culture. But this is tantamount to arguing for an essential axiom. That individual judgments are better accommodated constitutes evidence for the basic commitment.

Furthermore, this new essential axiom may account for individual judgments which the old did not. Consider a materialist and a theist with their contrasting world views. Could not both agree that human beings have human rights? Could not both substantially agree on what are those rights? But is it not conceivable that given one’s world view, one might construct a prima facie more reasonable or otherwise better explanation of why humans have rights and justification for respecting those rights than one might be able to construct given a contrasting world view? Might this not move an adherent of the other world view, at least in some way, to reconsider her world view commitments? That is, has the dialogue not taken a step toward the resolution of the disagreement through argument? Again, we are speaking quite generally here, surely could not a prima facie acceptable explanation of human equality in one culture on the basis of its world view commitments influence the ongoing argumentation in another culture whose world view commitments may not provide an equally prima facie  adequate explanation of human equality? Could not such ongoing argumentation lead to an increased convergence of points of view between the two cultures? At the least, entering such a dialogue may lead to a deeper understanding of one’s world view and a more mature commitment to it.

Surely, it is plausible that dialogues involving cross-cultural argumentation might lead to such an outcome. But such dialogues have a necessary condition – the participants must be genuinely open to valuing reasonableness. But need this always be the case? Our considerations here have not shown any reason to refuse to invite those with divergent world view commitments or indeed with any difference in viewpoint over significant, existential issues into a critical discussion. The question, of course, is whether they will accept the invitation. Satan certainly would not. If one’s world view denies that there can be evidence of a certain type, or that certain values are not genuinely positive but rather perverse, or claims that certain explanations which in open court might be judged best explanations are not viable at all, there may simply be nothing to say to that person in a critical discussion aimed at showing the reasonableness of one’s world view. Argumentation is limited by the willingness to enter into such dialectical exchanges. But for those who do accept the invitation, critical discussion offers a way of at least appreciating the reasonableness of others’ world views, and quite possibly of deeper understanding and refinement of one’s own. Issues of fundamental commitments, essential axioms, world-views are not then beyond the realm of argumentation. These claims are subject to support through argumentation where the recognitions of evidentiary relevance are independent of originating acts of faith. We see Fish’s skepticism of argumentation not justified on any level.

What then is the place of argumentation (and thus the importance of argumentation theory) for the present time with its deep cultural differences, which militants may seek to exploit, even violently. Such militants may be closed to entering a critical discussion. But this is not because their world view commitments and those whom they oppose are based on originating commitments which for all parties are arbitrary and immune to rational evaluation. Their refusal in no way shows that the invitation to inquiry was conceptually incoherent or critical discussion an impossibility. By contrast, if critical discussion is a genuine possibility, then there is at least one place in this pluralistic but currently increasingly polarized world where divergent cultures may meet to critically examine their differences in peace, where argumentation provides the framework for such meetings.

NOTES
[i] For our analysis of enthymemes and references to related literature, see our (2011), Chapter 7.
[ii] For our definition of interpretation as a type of statement and our distinction of the basic types of statements, see our (2005a, Chapter 5.2, especially p. 105).
[iii] The types of associated conditionals assumed parallels the types of warrants an argument may involve. For a discussion of these types, see our (2005b).
[iv] He realizes this unless, of course, his originating act sanctions circular inference.
[v]  Some argumentation theorists have found Toulmin’s notion of field problematic. In (2005b), we argue for replacing this notion with an epistermic classification. The points are still the same. Warrants can be backed, albeit in different ways, and different persons may develop different bodies of warrants.

REFERENCES
Eemeren, Frans H. van and Rob Grootendorst. (1984). Speech Acts in Argumentative Discussions. Dordrecht-Holland/Cinnaminson-U.S.A.: Foris Publications.
Eemeren, Frans H. van and Rob Grootendorst. (1992). Argumentation, Communication, and Fallacies: A Pragma-Dialectical Perspective. Hillsdale, NJ: Lawrence Erlbaum Associates, Publishers.
Fish, Stanley. (1996). “Why We Can’t All Just Get Along.” First Things 60, 18-26.
Freeman, James B. (2005a). Acceptable Premises: An Epistemic Approach to an Informal Logic Problem. Cambridge: Cambridge University Press.
Freeman, James B. (2005b). “Systematizing Toulmin’s Warrants.”  Argumentation 19 (2005), 331-346.
Freeman, James B. (2011). Argument Structure: Representation and Theory. Dordrecht: Springer.
Hitchcock, David (1985). “Enthymematic Arguments.” Informal Logic 7, 83-97.
Milton, John. (1962). Paradise Lost: A Poem in Twelve Books. Merritt Y. Hughes, ed.. New York: The Odyssey Press.
Peirce, Charles S. (1955). Philosophical Writings of Peirce.  Justus Buchler, ed. New York: Dover Publications.
Rawls, John. (1971). A Theory of Justice. Cambridge, MA: Harvard University Press.

1. Introduction
“Controversies are indispensable for the formation, evolution and evaluation of (scientific) theories, because it is through them that the essential role of criticism […] of scientific theories is performed” (Dascal 1998, p. 147). Of the many questions related to this claim, which we accept, we should like to focus on the question how present-day interactive digital media can be used as vehicles of public controversy in the sciences.

Historically, new media have often played a decisive role in facilitating public controversy. A case in point is the revolution in scientific communication caused by the introduction of scientific journals like the “Journal des Sçavans” or the “Acta Eruditorum” in the second half of the 17^th century. These journals appeared at relatively short intervals and provided the opportunity to report on one’s own research or, by writing reviews, to report and criticize the work of others, for scientists all over Europe to read and to respond to. These new media changed three important factors of scientific communication:
1. the spread of scientific information,
2. the speed of publication,
3. the amount of interactivity between scholars.

Maybe the most remarkable result of these changes was the opportunity provided for a multitude of lively public controversies in the Republic of Letters, which contributed to the confrontation and development of theoretical views and empirical research and thereby helped advance science in an amazing way.

Recent developments in digital technology have initiated changes in the practice of scientific communication which, arguably, are comparable to the 17^th century revolution in scientific communication.[i] What is remarkable is that factors similar to those three hundred years ago play a significant role in the use of recent new media, i.e. wide distribution, speed of publication, and a high degree of interactivity.

As observers of scientific communication today we are in the happy position to be able to follow the progress of evolving digital media and genres of communication in our own present time. This is what we are doing in a project on “Scientific information, critique and controversy in digital media”, which is being conducted at the University of Gießen (Germany) and which is funded by the VW Foundation.[ii] Our paper presents work done in the context of this project, focussing mainly on controversies in interactive digital formats like mailinglists, blogs, and open-review journals. As for our theoretical approach, we build on our earlier work in the pragmatics of controversies and on communication in the digital media (cf. Fritz 2008; 2010; forthcoming a; Gloning 1999; 2005).

2. On the attractivity and some problems of scientific controversies in interactive digital media
If controversies are considered an efficient motor of scientific progress, then it could be a measure of the success of the new digital science media, if these media encourage fruitful controversies. There is, however, so far no simple answer to the question if this is the case.

Generally speaking, there is an interesting tension between the fact that many scholars are quite reluctant to participate in controversies on the internet and the fact that we do find many attractive and worthwhile controversies in these formats. As for the reasons for this reluctance, scholars we asked mentioned the following, among others:
– Controversies are too time-consuming.
– Controversies can be harmful to your reputation.
– Collaborative efforts like the participation in controversies don’t pay out in terms of the academic reward system.
– Theoretical controversies are less useful than the collection and analysis of empirical data.

These and similar reasons seem to be obstacles to active participation in scientific controversies. Obviously, what is considered an obstacle differs according to subject or discipline. For example, open peer review has been practised in Physics and other sciences for about 20 years now, whereas Arts subjects still tend to stick to traditional reviewing of papers. This is an interesting point which we shall, however, not discuss in this paper.

In spite of these obstacles, many interesting controversies are conducted in digital formats. From what we have seen in our research so far, there are especially two types of contexts where lively controversies tend to arise. The first is topics and domains where scientific research and public interest meet, e.g. climate controversies or controversies on creationism and similar topics. The second context is reviews of scientific writings and reactions to such reviews. We shall briefly mention an example of the former type and then go on to summarize two case studies on controversies sparked off by reviews.

Discussions on topics on the borderline between science and politics and ideology are often quite animated and informative, there is, however, a tendency for ideological dogmatists and other destructive participants (so-called “trolls”) to intrude on and even to dominate such discussions, which makes them less attractive for “genuine” scientists. We should like to give an example of this kind of thread in the medium of blogs.

On July 30^th, 2008 a paper with the title “Dinosaurian Soft Tissues Interpreted as Bacterial Biofilms“ by T. G. Kaye and his collaborators appeared in PLoS ONE, an interactive open-access journal for the communication of peer-reviewed scientific and medical research.[iii] This paper was a critical reaction to earlier studies, which had claimed to have identified and isolated soft tissues from a 68 million year old fossil bone. On the day of its publication in PloS ONE, Tara C. Smith, an Assistant Professor of Epidemiology, summarized the article by Kaye and part of the earlier controversy on her own blog Aetiology and explained its main point to non-specialists.[iv]

This blog was commented upon in 20 postings within two days. Two of the postings are particularly interesting from our point of view, because they show part of the process which contributes to the wide distribution of contributions on the internet. The first is by Tom Kaye, one of the authors of the paper

Hello All,
Tom Kaye here from the paper. Since this seems to be the blog with the most activity, I will offer to answer any questions for the group.
Tom
Posted by: Tom Kaye | July 30, 2008 5:11 PM

The second one is by the owner of the blog, Tara C. Smith, who directly addresses Tom Kaye and mentions another blog, where there is a lively discussion on the same topic going on:
Hi Tom –
Thanks for stopping by! There’s also a good discussion over at Panda’s Thumb, where I cross-posted this. If you can ignore the trolls (the creationists etc.) there are some good
questions you may be able to respond to over there also.
Posted by: Tara C. Smith | July 30, 2008 5:56 PM

The relevant discussion on Panda’s Thumb, a scientific weblog on questions of evolution, comprises 122 comments within a fortnight.[v] Among these postings there are quite a number of serious, scientifically-informed contributions, to which the author answers in longish replies. But there is also at least one obvious anti-evolutionist, who introduces a fairly polemical tone. To this the author of the paper remarks: “I see there is the usual ID (i.e. Intelligent Design, GF) spam going on but if we can work around that I am willing to answer any reasonable questions”. So what we get on this blog is a mixed bag of serious discussion and facile polemics. And much of this is happening on the very day the Kaye et al. paper was published. So, whatever the merits of this discussion in terms of scientific progress, the author of the paper certainly received a remarkable amount of “attention space” (cf. Collins 2000, p. 38f.) for his research within a short period of time.

3. Reviews and replies
Now to the question of controversies sparked off by reviews. We shall give two examples from case studies from our project, one taken from a mailinglist and one from an open peer review journal.

3.1 A review and an ensuing controversy on a mailinglist
The first example consists of material from the Linguist List section on “book discussion” which we shall briefly present and analyse. The Linguist List is the biggest website for academic linguists, providing mailing lists for various sub-disciplines.[vi] The purpose of the book discussion section is presented as follows: “We strongly encourage discussion (including book authors if they so desire and their response is appropriate) of reviews. We do this because we feel the electronic medium allows us to provide a service that print sources cannot” (posting by the moderator in charge of reviews). A later notice by the moderator sounded even more inviting to authors: “What follows is a review or discussion note contributed to our Book Discussion Forum. We expect discussions to be informal and interactive; and the author of the book discussed is cordially invited to join in” (Andrew Carnie, in a posting of Oct. 3^rd, 2000).

This is the exact opposite of the principle that an author should not reply to his reviewer, which is still well established in scientific journals today, although historically, this is by no means necessary, as the early history of reviewing in the 17^th and 18^th centuries shows.[vii] So, in this respect, we are back to the exciting days of the late 17^th century!

We shall now give a short analytical summary of a controversy which took place few years ago, and which nicely shows the potential of the mailinglist format for this kind of exchange.[viii]

On July 3rd, 2002 Joybrato Mukherjee published on this list a review of the “Cambridge Grammar of the English Language” by Rodney Huddleston and Geoffrey K. Pullum, published in 2002 by Cambridge University Press. The following controversy consisted of three further contributions, a response to this review by Pullum, posted on July 15^th, a reply to this response by Mukherjee on July 20^th, and a final reply by Pullum on July 22^nd. Looking at these dates, we already notice one characteristic feature of this kind of exchange, namely, the relative speed of reaction in the interactive process.

The content of the review can be described as follows: Mukherjee starts off by praising the “admirable achievement and the monumental quality of this volume“ and then goes on to give a survey of the content of the chapters of the grammar. After these largely descriptive passages, Mukherjee turns to a critical evaluation. His main points of criticism concern the presumed fact that this grammar is mainly based on one grammatical model, i.e. Generative Grammar, and that it is not “a genuinely corpus-based description of English”. There are also some minor objections, which we shall not mention here.

In his response to this review, Pullum starts by mentioning Mukherjee’s two main objections: “He criticizes [the grammar] for not being corpus-based, and for adopting analyses on grounds of dogma rather than evidence.” He then criticizes Mukherjee for failing “to show respect for textual evidence”, the latter remark being a classic tit-for-tat move. He then asserts that “all his negative criticisms of [the grammar] rest on false claims” and decides to “offer a brief response to half a dozen especially egregious ones”. He now numbers his objections from 1 to 6 and deals with each one in detail. (This practice of numbering objections is a classic procedure, which goes back at least to the 16^th century.)

In his rejoinder, Mukherjee first accuses Pullum of presenting his reviewer as “someone who lacks even basic reading skills” and announces his intention to correct this picture. He then takes up all Pullum’s objections and deals with them point by point in the order presented by his opponent. This procedure is again a traditional pattern of topic management in scientific controversies. Mukherjee’s rejoinder, which amounts to 3698 words, includes the discussion of conceptual problems, theoretical arguments against Pullum’s position, and the giving of counterexamples and references. So this contribution to the controversy is very much in the tradition of scientific writing as we find it in books and articles, but not normally in a defence of a review.

In the final contribution to the controversy, Pullum uses a very interesting strategy, which consists in claiming that “despite the trappings of squabble and a charge of “strangely offensive tone”, much agreement emerges on matters of fact”. He then goes on to enumerate 10 points of agreement, which he briefly deals with in the course of his posting. Looking at these points closely, one realizes that his presentation of “agreement” mainly serves to assert his own position in the controversy. At one point, he admits that in the discussion he “took the liberty of a little ad hominem dig in the ribs against Mukherjee”. And finally, he acknowledges that “Mukherjee’s review made numerous positive statements and generous remarks”. So, in spite of a polemical note here and there, politeness and a factual tone prevail.

Now, how do we evaluate the quality and the usefulness of this controversy?  One would probably agree that this exchange of arguments came up to the standards expected of scientific discussions and that it contributed to the clarification of the positions involved. For the opponents, the discussion provided an opportunity to broadcast their views, and for novice grammarians and non-specialist linguists it provided an introduction to a major conflict in present-day grammar writing between theory-based and corpus-based conceptions. Considering in addition the comparative speed of publication and its wide distribution, this type of review-cum-discussion on mailing lists or blogs can certainly be considered a useful addition to the formats of scientific dialogue. One of the most interesting features of this interactive procedure is that it causes changes in the roles of reviewer and author, as both have to envisage a course of events in which they might become participants in a serious controversy.

3.2. Public peer review
A different type of communication between reviewers, authors and the scientific public can be found in open peer review journals, which aim to make the reviewing process for research papers more transparent and, in some cases, publicly accessible. Among the new open access journals we find different versions of the reviewing process, which vary as to the amount of interactivity and transparency in the different phases of the reviewing and publication process.

A fully developed interactive reviewing process was introduced in 2001 by Atmospheric Chemistry and Physics (ACP), “an international scientific journal dedicated to the publication and public discussion of high quality studies investigating the Earth’s atmosphere and the underlying chemical and physical processes”.[ix] “Atmospheric Chemistry and Physics has an innovative two-stage publication process involving the scientific discussion forum Atmospheric Chemistry and Physics Discussions (ACPD). […] In the first stage, papers that pass a rapid access peer-review are immediately published on the Atmospheric Chemistry and Physics Discussions (ACPD) website. They are then subject to Interactive Public Discussion, during which the referees’ comments (anonymous or attributed), additional short comments by other members of the scientific community (attributed) and the authors’ replies are also published in ACPD. In the second stage, the peer-review process is completed and, if accepted, the final revised papers are published in ACP.”[x]

I shall now sketch some observations on one of the most lively controversies conducted on the ACPD discussion forum, the discussion on a paper by A. M. Makarieva and two collaborators “On the validity of representing hurricanes as Carnot heat engines” (Atmos. Chem. Phys. Discuss., 8, 17423-17437, 2008). After the preliminary reviewing process, the paper was published as a “discussion paper” on Sept. 19^th, 2008. As they state in their abstract, the authors “argue, on the basis of a detailed critique of published literature, that the existing thermodynamic theory of hurricanes, where it is assumed that the hurricane power is formed due to heat input from the ocean, is not physically consistent, as it comes in conflict with the first and second laws of thermodynamics.” They claim, in fact, that this theory makes a hurricane a perpetuum mobile. In the second part of their paper they outline an alternative explanation based on the description of an “atmospheric process occurring at the expense of condensation of water vapour that creates a drop of local air pressure”. It is interesting to see that in the following discussion the main point of attack is the challenge presented by the authors to the widely accepted “standard theory” of hurricane formation.

For reasons of space, we cannot here go into details of this controversy, which consists of 35 postings, taken all together.[xi] We should, however, like to comment on a few aspects of the external structure of the controversy, which can be seen in the following survey given in the ACPD archive:[xii]

AC: Author comment  RC: Referee comment  SC: Short comment  EC: Editor comment

AC S7325: ‘Response to preliminary criticisms’, Anastassia M. Makarieva, 20 Sep 2008
RC S7915: ‘Review ‘, Anonymous Referee #1, 03 Oct 2008
AC S7947: ‘Response to Review of Referee 1’, Anastassia M. Makarieva, 04 Oct 2008
RC S8170: ‘Follow-up’, Anonymous Referee #1, 12 Oct 2008
AC S8193: ‘Response to Follow-Up by Referee 1’, Anastassia M. Makarieva, 13 Oct 2008
AC S9182: ‘Final Response: Heat Release to Space’, Anastassia M. Makarieva, 16 Nov 2008
SC S7609: ‘Latent work’, Anastassia M. Makarieva, 29 Sep 2008
SC S8318: ‘Motion from condensation’, Semen Sherman, 17 Oct 2008
AC S8340: ‘Latent work: Convective potential energy’, Anastassia M. Makarieva, 18 Oct 2008
SC S8164: ‘The novel hurricane physics’, Andrei Nefiodov, 11 Oct 2008
RC S8531: ‘Review ‘, Anonymous Referee #2, 25 Oct 2008
AC S8904: ‘Condensation as Air Circulation Driver’, Anastassia M. Makarieva, 10 Nov 2008 RC S9081: ‘Extraordinary novel atmosphere physics’, Anonymous Referee #2, 13 Nov 2008
SC S11826: ‘Considerations of turbulent friction’, Anastassia M. Makarieva, 22 Mar 2009
RC S8627: ‘This paper is incoherent’, Anonymous Referee #3, 29 Oct 2008
AC S8635: ‘Response to Referee #3’, Anastassia M. Makarieva, 30 Oct 2008
SC S8669: ‘The Sun does not orbit around the Earth.’, Paulo Nobre, 30 Oct 2008
SC S8916: ‘paper contains bad physics’, Antoon Meesters, 10 Nov 2008
AC S8923: ‘Bad physics: Latent heat does not warm’, Anastassia M. Makarieva, 10 Nov 2008
SC S8979: ‘latent heat in the atmosphere’, Antoon Meesters, 11 Nov 2008
AC S8998: ‘Latent heat is irrelevant’, Anastassia M. Makarieva, 12 Nov 2008
AC S8931: ‘On carelessness and responsibility’, Anastassia M. Makarieva, 10 Nov 2008
SC S9060: ‘dissipative engine etc.’, Antoon Meesters, 12 Nov 2008
SC S8953: ‘The “subtle” issue of perpetuum mobile’, Semen Sherman, 11 Nov 2008
AC S11647: ‘Comment on the dissipative heat engine’, Anastassia M. Makarieva, 15 Mar 2009
AC S9342: ‘Final Response to Dr. Meesters’, Anastassia M. Makarieva, 20 Nov 2008
AC S11254: ‘Final Response: List of Revisions’, Anastassia M. Makarieva, 14 Feb 2009
AC S11260: ‘Revised manuscript, part I’, Anastassia M. Makarieva, 14 Feb 2009
AC S11275: ‘Revised manuscript, part II’, Anastassia M. Makarieva, 14 Feb 2009
AC S12153: ‘Appeal to the ACP executive committee’, Anastassia M. Makarieva, 02 May 2009
EC S12168: ‘Editor Report’, Peter Haynes, 04 May 2009
EC S12406: ‘Final Editor Comment (ACP Exec. Editors)’, Ulrich Pöschl, 14 Oct 2009

Apart from the authors and three reviewers, there are four more participants in this controversy. Three fellow scientists post short comments in which they support the views of Makarieva et al. A fourth scientist, a Dutch physicist and meteorologist, posts a longish comment, in which he puts forward a number of objections against Makarieva’s paper and gives arguments in favour of the standard theory. This posting is answered in detail by Makarieva, which leads to a mini-discussion within the total controversy. It is this thread of postings which shows to advantage the potential of the ACPD system for involving specialists outside the circle of reviewers in the open reviewing process.

Of the many interesting aspects of this controversy we shall now pick out one point of conflict which highlights some problems and principles of open peer review. On Oct. 29^th, the third reviewer posts his first public comment and asserts that he finds “this paper to be incoherent at the least” and that it “is not worthy of publication in any respectable journal”. He furthermore states that the strong criticism of the classical theory was not well-founded and claims that much of the Makarieva paper was incomprehensible and what he did understand was wrong. He concludes by repeating his harsh judgement.

By this highly polemical posting, the third reviewer creates a rather difficult position for Makarieva and colleagues, who still count on having their paper published. In their reply of Oct. 30^th they use a double strategy of attempting to convince the reviewers of the well-foundedness of their criticism and of reflecting on the course of the discussion itself. They start out with a polite move, appreciating the call for serious justification of their criticism. They then go on to point out their arguments and where they are given in detail and also expand on some of these arguments. We will skip this bit, which contains a lot of technical detail, and go to the last part of their reply, which is particularly interesting, as it concerns the style of the controversy and fundamental principles of open peer review:
“Finally, we would like to note that, in our view, the open discussion platform of the EGU journal sets up a new and high ethical and cultural standard of the peer review process. In this context, statements like “this paper is not worthy of publication in any respectable journals” should perhaps be viewed as atavisms of the background private communication between the editor and referee during conventional close review process. When such statements are made in open public discussion potentially read by hundreds of people, especially in the view that the referee cannot follow “much of the argument here”, they can be classified as a public assault to both the authors as well as to all those discussion participants who sign their names under very different opinions as well as to the ACPD journal itself (who did publish the paper).

Moreover, in our view, the above statement of referee 3 goes against the journal’s interest not only in its form, but also in its essence. We believe that the main target of this discussion is to reveal the scientific truth. The discussion paper is citable, covered in Scopus and available for analysis. Indeed, we come up with a rare claim that a framework published in high-profile journals is based on the concept of a perpetual motion machine and is fundamentally incorrect. Our arguments are all here. In our view, if our paper were published in ACP, then the responsibility to respond to our critique would go to the author of the criticized framework, as the normal practice in scientific literature goes. If, on the other hand, the ACP declined our paper for publication in the second stage, as recommended by Referee 3, future readers of this discussion would ultimately decide whether or not the journal actually signed its official name (while Referee 3 remaining anonymous) among the defendants of perpetuum mobile and against a new approach to hurricane physics. In any case, however, we believe that this discussion has a very substantial value. We are very grateful to the journal for letting us express our views on its pages.”

This is a remarkable document, touching on various basic aspects of open peer review, e.g. politeness and fairness principles, the responsibilities of the participants, the anonymity of reviewers, the burden of proof in scientific argument, and the question of who is “judge of controversies” in science.[xiii] It shows that many of the fundamental principles of scientific discourse acquire particular relevance and salience in public peer review and public digital controversy in general. This is especially true of principles guarding against face-threatening acts.

It is worthy of note that in the end the paper was not accepted for publication, as the objections formulated by two of the three referees appeared so fundamental to the managing editor that, on close reflection, he did not believe that the paper in its present form reached the standards required for publication and that he did not see “a straightforward route to changing it to make it publishable” (editor’s report). However, the chief executive editor considered this case exceptional enough to decide to re-assess the judgement of the referees and the managing editor some months later, bringing in two additional referees, and to give a final statement on the procedure and its results. In this final statement, he writes: “I am not a specialist in atmospheric dynamics and meteorology, and I found the exchange of arguments between authors and referees interesting and challenging. In this regard, I would like to express my appreciation for the clear formulation and mathematical precision of the line of arguments and comments of Dr. Makarieva and co-authors. – After all, however, I have come to share the specialist referees’ concerns that crucial assumptions underlying the arguments, comments and manuscript of Makarieva et al. appear not to be justified.” Obviously still feeling some misgivings about the outcome of the reviewing process, he finally reflects on the principles of open peer review, which, whatever the outcome of the reviewing process, are meant to secure a high degree of transparency. One of his final remarks is as follows: “In the present case, free speech and public documentation have already been achieved by publication of the discussion paper in ACPD, and Makarieva et al. have also taken the opportunity of publishing a revised version of their manuscript in the form of interactive comments in ACPD. As mentioned above and detailed on the ACP web pages, the discussion paper as well as the interactive comments will remain permanently archived, accessible and citable.”

Generally speaking, this kind of exchange shows the potential and scope for fruitful public scientific discussion in this type of reviewing process. As for the different participants, this type of interaction provides new opportunities, but it also poses new communicative tasks. Reviewers have to keep in mind that their reviews will be publicly available for criticism not only by the authors, but also by the relevant scientific community at large. This calls for a high level of rational argumentation and commits the reviewers to principles of politeness and objectivity. So, in a way, reviewing is harder in this kind of framework. And, of course, having to answer objections to your review can be hard work. This might be one of the reasons why finding a sufficient number of qualified reviewers is one of the major problems of open peer review. Authors have the opportunity to have their work closely scrutinized before it is finally put in print and they have the chance to receive attention – once their paper has cleared the hurdle of access review –, whether their paper is finally accepted or not. On the other hand, they have the obligation to answer objections in public within reasonably short time, which can be quite a challenge and possibly a problem for their reputation. For authors of short comments this option provides the chance to take part in a public scientific discussion without having to produce a paper of one’s own, and for the lookers-on it provides the opportunity to recognize conflicting views and to observe the arguments for these views being presented in actual performance.

So, to sum up, these forms of interactive reviewing seem to present a healthy challenge to the participants in the reviewing process. As yet it is mainly in the field of natural science that open peer review has been adopted. We shall see if in the future the arts and humanities will follow suitable.

4. Conclusion
We should like to conclude with some reflections concerning the potential of digital formats for fruitful scientific controversy and the conditions under which digital-format controversies will be productive.

Generally speaking, speed of publication and the wide distribution of postings, which are both characteristic properties of communication in digital formats, seem to be ambivalent factors that can be either favourable or unfavourable to high-quality scientific controversies. Speed of publication, including speed of reaction, often creates a certain “flow” of interaction, which may stimulate a lively discussion. On the other hand, rash replies increase the risk of injury that is always present in controversies. Therefore, members of lists or commenters on blogs are sometimes advised to count to ten, before they hit the reply key. Wide distribution and open access may be helpful in attracting qualified disputants, but it may also attract unqualified and disruptive participants. So balancing these factors seems to be an important task of the respective communities.

From what we know today, the following three conditions play an important role in generating productive controversies:
1. Prospects for useful controversies seem to be particularly good in fairly close-knit scientific communities with a reasonable number of active participants. It is in such specialist communities that the motivation to actively contribute to discussions and the ability to deal rationally with conflicting views appear to be highest. This observation seems to be in conflict with the view that open access for a wide scientific public is a strong point of digital formats. But in practice, it is often a small group of persons who dominate the actual interaction, quite independent of the large number of “lurkers” that may passively participate.
2. A second condition of good controversy is close attention to topic management. Initiating relevant and attractive topics and keeping a discussion on track without restricting creative developments is an important task of the contributors. In many cases this is accomplished naturally and without an extra effort by participants, but rambling or disruptive postings are always a risk to be aware of. The observation that it is often reviews which spark off good discussions is probably connected to the fact that both authors and reviewers are genuinely motivated to defend their point of view and to the fact that the book or article under review provides a natural topic focus.
3. Finally, it is often the moderators of mailinglists or the owners of blogs who contribute to the development of good controversies on their lists or blogs by suggesting salient topics, by organizing round tables or blog carnivals, and by generally trying to sustain a well-organized procedure by which the “vices of confused disputes” (Leibniz 2006: 1-6) can be avoided. So being organized by an active and responsible moderator or owner can be a decisive factor for the success of a digital format in facilitating fruitful controversy.

Certainly, these or similar conditions are not exclusively relevant to digital formats, but may also play a role in any format of scientific communication. However, under the specific conditions of digital scientific communication, which we mentioned above, they acquire particular salience.

Trying to weigh up the potential and the risks of digital scientific controversy, we seem to face a similar situation as the 17^th-century pioneers of research journals we mentioned at the beginning of our paper. It remains to be seen, if, in the long run, the members of the scientific community will avail themselves of the potential of the new formats with the same enthusiasm as their 17^th-century forebears did.

NOTES
[i]  Some authors have emphasized the influence of these new media on recent developments of science by using the expression “cyberscience” (e.g. Nentwich 2003).
[ii]  More details on the project can be found on the project website: Website: http://www.zmi.uni-giessen.de/projekte/zmi-isteilbereich4.html
[iii]  URL: http://www.plosone.org/home.action (25.02.2010)
[iv] URL: http://scienceblogs.com/aetiology/ (25.02.2010)
[v] URL: http://pandasthumb.org (25.02.2010)
[vi]  URL: http://linguistlist.org/ (25.02.2010)
[vii]  For the history of critical reviews and replies to reviews („anti-critique“) in early scientific journals, cf. Habel (2007).
[viii]  The review and the responses are available on the Linguist List Review Archives (issue numbers: 13.1853, 13.1932, 13.1952, 13.2005) (10.07.2010).
[ix]  For some of the ideas behind the introduction of public peer review, cf. Pöschl (2010), an article by the chief executive editor of ACP.
[x]  URL: http://www.atmospheric-chemistry-and-physics.net/home.html (25.02.2010)
[xi] A detailed analysis of this controversy is presented in Fritz (forthcoming b).
[xii] URL: http://www.atmos-chem-phys-discuss.net/8/17423/2008/acpd-8-17423-2008-discussion.html (05.07.2010). The dates of the individual postings show, among other things, how quickly the authors reacted to the various queries and objections.
[xiii] For some early reflections on “the judge of controversies”, cf. Leibniz (2008), Ch. 8.

REFERENCES
Collins, R. (1998). The sociology of philosophies. A global theory of intellectual change. Cambridge, Mass.: The Belknap Press of Harvard University Press.
Dascal, M. (1998). The study of controversies and the theory and history of science. Science in Context, 11, 147-154.
Fritz, G. (2008). Communication principles for controversies: A historical perspective. In: F. H. van Eemeren, B. Garssen (eds.) Controversy and Confrontation: Relating Controversy Analysis with Argumentation Theory. Amsterdam/ Philadelphia: John Benjamins, 109-124.
Fritz, G. (2010). Controversies. In: A. H. Jucker, I. Taavitsainen (eds.) Historical  Pragmatics. Handbook of Pragmatics, Vol. 8. (pp. 451-481). Berlin/New York: de Gruyter.
Fritz, G. (forthcoming a): Scientific dialogue in digital interactive media. To appear in: S. Cantarini, F. Hundsnurscher (eds.). Dialogue: Studies in memory of Sorin Stati. Bologna: Patron. (Manuscript download: http://www.zmi.uni-giessen.de/projekte/zmi-isteilbereich4.html)
Fritz, G. (forthcoming b): Wirbelstürme im digitalen Open-Peer-Review-Verfahren. Die Makarieva-Kontroverse in “Atmospheric Chemistry and Physics” (2008/09) – eine Fallstudie. In: T. Gloning, G. Fritz (eds.): Kommunikationsformate und ihre Dynamik in der digitalen Wissenschaftskommunikation. Reihe “Linguistische Untersuchungen”. Gießen: Gießener Elektronische Bibliothek.
Gloning, T. (1999). The pragmatic form of religious controversies around 1600: A case study in the Osiander vs. Scherer & Rosenbusch controversy. In: A. H. Jucker, G. Fritz & F. Lebsanft (eds.). Historical Dialogue Analysis (pp. 81–110). Amsterdam/ Philadelphia: John Benjamins.
Gloning, T. (2005). Early Modern controversies and theories of controversy: The rules of the game and the role of the persons. In: P. Barrotta, M. Dascal (eds.): Controversies and Subjectivity. (pp. 263–281). Amsterdam/ Philadelphia: John Benjamins.
Habel, T. (2007): Gelehrte Journale und Zeitungen der Aufklärung. Bremen: Edition lumière.
Leibniz, G. W. (2008): The art of controversies. Translated and edited, with an introductory essay and notes by M. Dascal with Q. Racionero and A. Cardoso. Dordrecht: Springer.
Nentwich, M. (2003): Cyberscience. Research in the age of the internet. Vienna: Austrian Academy of Sciences Press.
Pöschl, U. (2010): Interactive open access publishing and public peer review: The effectiveness of transparency and self-regulation in scientific quality assurance. International Federation of Library Associations and Institutions, 36 (1) 40–46.