ISSA Proceedings 1998 – Linguistically Sound Arguments
The centuries-long discussion as to what constitutes “good” argument has often found supporters and opponents on the basis of the standards selected to evaluate argument. Ancient standards of technical validity have been the subject of some twentieth-century scrutiny. No issue is more fundamental to the study of argumentation than the question of what constitutes good argument. Our legitimacy as critics, practitioners and teachers of argumentation rests upon our ability to evaluate, construct and describe good arguments. Historically, argument scholars have relied primarily upon formal standards borrowed from the field of logic to provide necessary evaluative criteria. In the latter half of this century, however, those criteria have increasingly been attacked as being inappropriate or, at least, insufficient for the study of both public and personal argumentative discourse. Stephen Toulmin has suggested we replace the mathematical model of argument with one from jurisprudence, thus focusing on the soundness of the claims we make, especially as we use argument in “garden variety discourse.”(Toulmin, 1958). Other theorists quickly followed Toulmin’s lead.
1. Recent Interpretations of Good Argument
While a few theorists (Willard, 1979) have gone so far as to reject logical standards, most others continue to recognize their usefulness as a part of broader schemas for evaluation of argument. Toulmin’s dissatisfaction with the rigidity and formalism of logic led him to propose a more open and flexible model of argument and to suggest that the evaluation of arguments involves the application of both traditional field invariant standards and previously overlooked field specific standards (Toulmin, 1958). Perelman and Olbrechts-Tyteca have advanced the concept of the universal audience composed of critical listeners, which presumably restrains advocates from making spurious arguments. At the same time, they suggest we consider adherence as the goal of argument, a focus on the intersection of psychological effects and logical strength (Perelman and Olbrechts-Tyteca, 1969). Drawing on the work of earlier scholars, McKerrow describes a good argument as one which provides “pragmatic justification (McKerrow, 1977). This interpretation places emphasis on the “rational perusal of arguments” by an audience in a dialectic-like relationship. Farrell interprets validity in terms of “soundness” of a rhetorical argument. An argument is sound if it conforms to three conditions:
1. is addressed to an empowered and involved audience,
2. conforms to the consensual standards of the specific field, and
3. is consistent with social knowledge (Farrell, 1977).
Zarefsky defines good argument as one that is “reasonable,” and one is reasonable if “the form of inference is free of obvious defects, and the underlying assumptions of the argument are shared by the audience” (Zarefsky,1981:88).
Collectively, these authors and others suggest that good arguments are ones that have, at least, some claim to rationality and are based upon premises and standards acceptable to the specific audiences being addressed. While these conditions serve as minimal standards for good argument, they are, in our judgment, incomplete and lacking in explanatory power. What is missing from current analyses is a consideration of the role of language. Careful language usage is necessary for the construction of sound arguments, and effective language is the key to persuasive argumentation. We define a good argument as one that is linguistically sound. The term “linguistically sound” is intended to encompass three conditions. A linguistically sound argument:
1. conforms to the traditional field invariant standards of inductive and deductive argument,
2. is based upon data appropriate to the audience and field, and
3. is expressed in language that enhances the evocative and ethical force of argument.
In the sections that follow, we will demonstrate how each of these conditions is linguistically based and how a linguistic perspective helps to explain the strength of the argument.
2. Field Invariant Standards
Even a cursory examination of argument suggests a close relationship between language and argument. It is through language that we describe relationships and create meaning about the world around us. Concepts such as correlation and causation allow us to perceive relationships differently than was possible before we had appropriated these methodological terms. We may have an intuitive sense of justice and love, but our ability to differentiate them occurs through language. Thus, language is the means by which we bridge the gap between the complex and confusing world of our senses and a more ordered world of meaning.
In his thoughtful essay, “Argument as Linguistic Opportunity,” Balthrop examines argument from a linguistic perspective and establishes a strong relationship between language and discurive reasoning. Discursive reasoning itself arises in discourse and shares its characteristics: that is, it posits relations both syntactically and semantically and through the fundamental representativeness of linguistic symbols. Second, discursive reasoning is sequential – for without sequence, verbal expression cannot exist. It is from such insights that Langer observed in Philosophy in a New Key, “the laws of reasoning, our clearest formulation of exact knowledge, are sometimes known as ‘laws of discursive thought.” If the symbolic function of argument is reason-giving or presenting justification, then that function is accomplished through discursive means – for reason giving requires analysis beyond mere expression. And, in the practical world of both the naive and the more sophisticated social actor, such analysis is usually conducted linguistically (Balthrop, 1980: 190).
Thus language becomes the key to discursive reasoning, and is central to the whole activity of reason giving. Balthrop goes further to argue that linguistic forms reflect how people think – at least at the deep structure level. He continues:
The subject-predicate structure for human thought may, in fact, be universal. Langer concludes that “to all speakers of Indo-European languages the classical syllogism seems to be a logic of ‘natural inference,’ because they speak and think in subjectpredicate forms.” Izutsu goes one step further contending that “far from being a peculiarity of Western thought /predicatesubject thought/ seems to be normal and universal wherever the human mind has attained a certain level of logical thinking as far, at least, as it is carried on by means of verbal symbols” (1980:195).
An understanding of the relationship between language and argument is important because it explains why the traditional field in-variant standards of inductive and deductive argument reveal potential problems in the thinking process. Even if the traditional standards are not a perfect reflection of the ways in which experience, language, and thought are related, no one has yet provided more useful tests. Although some may argue that Toulmin’s concept of field dependent standards makes traditional invariant standards irrelevant, it is well to remember that Toulmin, himself, did not propose field variant as a substitute for field invariant standards. Moreover, research to date has tended to reveal differences among fields only in the importance assigned to particular forms and standards of argument rather than in the forms and standards themselves. Perelman and Olbrechts-Tyteca’s concept of a universal audience is too abstract to be of much practical use for either the construction or criticism of arguments. And even Fisher’s concepts of narrative probability and narrative fidelity are only more generalized, and therefore, less analytical, forms of the traditional standards for evaluating arguments.
Thus, the field invariant standards of argument are an important component of a linguistically sound argument. They are grounded in our language and thought structures; they are supported by historic experience, and alternative standards seem to be insufficient. As Zarefsky concludes, reliance on these standards “in the past has led to satisfactory results far more often than not” (Zarefsky, 1980:88).
3. Data Appropriate to the Audience
The second condition for a linguistically sound argument is that the data must be appropriate to the audience and field. The audience has always been central to rhetorical theory so that Toulmin’s concept of field invariant standards of argument has been readily embraced by rhetorical scholars. Much of the literature of both classical rhetorical theory and contemporary field theory emphasizes the need for advocates to build their arguments on premises that are shared by their audiences. Bitzer’s “revisitation” of the enthymeme grounds his analysis in what the rhetor shares with his or her audience (Bitzer, 1959). So much importance is placed on shared assumptions that it sometimes appears that audiences can only be addressed on subjects they already believe in. What is often not discussed, however, is how an advocate can proceed if her basic assumptions differ from those of her audience. An examination of the role of language in argument is helpful in this regard.
Language can be used to create a greater harmony of beliefs than might otherwise exist. The ambiguous nature of values and the abstract language used to identify them make it possible to minimize differences and maximize agreement through careful conceptual choices. Kenneth Burke’s description of how dialectical terms (terms of opposition) may become transcendent (or terms of union, god terms) is a good illustration of this process (Burke, 1945). In recent years, politicians have regularly assumed that they and their audiences share a comon commitment to equal opportunity. Although most American audiences probably believe in equal opportunity at some level, such a belief does not translate into a common commitment to affirmative action; nor is a belief in affirmative action the same thing as a belief in racial and gender quotas. Thus, the ability to identify a common assumption and to link that assumption to an audience may depend in large part in the language of identification employed.
Not only are our beliefs abstract, but our belief systems encompass many different assumptions that exist in some loose hierarchy of values. This multiple, hierarchical nature of premises provides an additional opportunity for using language to establish a common ground. A linguistic bridge that embraces multiple beliefs can sometimes create a common ground out of conflicting assumptions. President Kennedy’s concept of a Peace Corps created such a linguistic bridge. The Peace Corps’ concept incorporated elements of economic assistance, service opportunities for young and elderly persons, and greater American involvement in foreign nations.
While the community service aspect of the program had relatively broad appeal, the ideas of increased foreign spending and greater U.S. involvement in the problems of third world nations were not popular with large segments of the American public. Kennedy’s labeling of the program as the Peace Corps allowed him to embrace all of these values and minimize resistance by linguistically identifying it with the higher, and more encompassing, shared value of peace. Premises are, of course, not the only form of data. When the shared assumptions of speaker and audience are insufficient and need to be built upon, evidence is required. The amount and type of evidence needed depends upon the expectations of the specific field and audience. But even within those constraints, language factors can significantly affect the impact and acceptability of that evidence.
When a range of expert testimony is available, the author’s language should be a fundamental consideration in deciding which source to rely on. The language used in the evidence should be free of offensive references. Currently, evidence which relies on “he” as a pronoun for persons in general may function to alienate certain audiences. In addition, the language should be appropriate to the level and background of the audience, and it should enhance the emotional and ethical appeal of the argument. Similarly, even statistical evidence is frequently difficult for audiences to comprehend so that special attention should be given to explaining and interpreting its meaning. For general audiences, the use of non-technical terminology is especially important. Whether data of fact or opinion, language functions centrally in both creating understanding of evidence for an audience and shaping audience attitudes toward that data.
4. Enhancing Emotional and Ethical Force
A third condition for a linguistically sound argument is that it be expressed in language that enhances the argument’s emotional and ethical force. The two preceding conditions of a good argument have generally been recognized by other authors, although they have focused less attention on the linguistic dimensions of these standards. The third condition of argument, however, has been largely overlooked as a positive element of argument. Logicians have generally viewed language as a negative factor in argument. Many of the logical fallacies, for example, are based upon language problems or upon unacceptable emotional or ethical appeals. Much of the rhetorical discussion of style has viewed it as an artistic adornment that functions to enhance effect but is largely unrelated to argument.
It is not our purpose here to disagree with specific categories of logical fallacies. We recognize that language can be misused and that the substitution of emotion or appeals to authority for reasoned argumentation is inappropriate. Nor do we wish to devalue the artistic dimensions of rhetoric. Rather it is our position that language is not only inherent to the argument process, but that an understanding of its proper role resolves the tension between the standards of logical validity and audience effectiveness.
Alan Gross and Marcelo Dascal in their essay “The Question of the Conceptual Unity of Aristotle’s Rhetoric” argue that in the Rhetoric inference (argument) is intimately related to language and style as well as to ethos and pathos. They describe Aristotle’s theory of language and style in the following terms:
Though little more than a sketch, Aristotle’s theory of style and arrangement is clearly cognitive in that it depends on the inferential abilities of particular audiences. Style is both a level at which discourse is pitched (in modern linguistics register) and a set of semantic, syntactic and prosodic variants within that register. In the former sense, a particular style is appropriate if it is proportional to situation and subject matter; in Aristotle’s words, “the lexis will be appropriate if it is … proportional /analogon/” (3.7.1). The mathematical analogy is exactly right; it emphasizes the close fit between a rhetorical situation and its verbal response (Gross and Dascal, 1998: 9).
In another passage, Gross and Dascal elaborate on Aristotle’s theory of emotion: …….. with Aristotle’s theory of emotions, a cognitive theory in which inference plays a central role ….. an audience experiences an emotional state when the necessary and sufficient conditions of that state have been met. Beliefs that speakers instill in audiences can never guarantee their anger. It certainly helps when audiences are, as Aristotle says, “irascible and easily stirred to anger” (2.2.10). Nevertheless, since the belief that one has been belittled or insulted is a necessary condition for the presence of this emotional state, speakers can stimulate anger by increasing inferential likelihood of that belief. Equally, speakers can dissipate anger by decreasing that likelihood. Inference to a articular belief or set of beliefs is a necessary condition of each emotion with which Aristotle deals – fear, shame, kindliness, pity, anger, friendship and their opposites (1998:9).
In his classic article on Aristotle’s enthymeme, James McBurney makes much the same point concerning how the forms of proof in Aristotle – ethos, pathos, and logos –relate to the dominant deductive and inductive forms of argument, the enthymeme and the example.
Rather than viewing the enthymeme and example as derivative of logos alone, he depicts both forms of argument as a product of the possible interaction of ethos, pathos, and logos. Hence the appeal to emotion, the possible instrument of style, such as the metaphor, or the character of the speaker may all interrelate in the production of an enthymeme. In this sense, the distinction between between language and argument may disappear, even in Aristotle (McBurney, 1936).
Even without an elaborate analysis of the cognitive dimensions of particular figures of speech such as those found in Aristotle’s Rhetoric, it is possible to demonstrate with references to familiar examples the evocative force that appropriate language gives to an argument. In his “House Divided” speech Lincoln used a powerful metaphor to express the fundamental claim of his speech.
“A house divided against itself cannot stand.” I believe this government cannot endure permanently half slave and half free. I do not expect the Union to be dissolved – I do not expect the house to fall – but I do expect will cease to be divided. It will become all one thing or all the other (Peterson, 1954:491).
Lincoln’s metaphor was not a mere rhetorical flourish. It was, rather, an integral part of his proof, and functions as a good example of metaphor as enthymeme. At that point in United States history, families were literally being torn apart over the issue of slavery so that the reference to a “house divided” served both as a appropriate metaphor and as compelling evidence of the crisis facing the nation.
William Faulkner’s speech accepting the Nobel Price offers a different, perhaps even more moving example, of how language enriches and empowers argument:
I decline to accept the end of man. It is easy enough to say that man is immortal simply because he will endure; that when the last ding-dong of doom has clanged and faded from the last red and dying evening, that even then there will be one more sound: that of his puny, inexhaustible voice, still talking. I refuse to accept this. I believe that man will not merely endure: he will prevail. He is immortal, not because he alone among creatures has an inexhaustible voice, but because he has a soul, a spirit capable of compassion and sacrifice and endurance. The poet’s, the writer’s duty is to write about these things (Faulkner, 1954: 815-16).
Faulkner’s argument is a simple one, but it is the imagery, the language of his imagination which gives the argument its ethical and emotional force.
In the terminology of the ancient Greeks, logos is not necessarily separate from ethos and pathos. Through the effective use of language these three forms of proof become united to form a linguistically sound argument.
A focus on language as the primary instrument of argument suggests that three necessary conditions exist for good argument. This paper explores the role of language in field invariant standards, how language functions in selecting and presenting data appropriate to the audience, and how language can enhance the emotional and ethical force of argument.
Balthrop, B. (1980). Argument as linguistic opportunity: A search for form and function. In: J. Rhodes and S. Newell (Eds.,) Proceedings of Summer Conference in Argumentation. (Pp. 184-213), Salt Lake City: SCA/AFA.
Bitzer, L. (1959). Aristotle’s Enthymeme Revisited. Quarterly Journal of Speech 45, 399-408.
Burke, K. (1945). A Grammar of Motives. Berkeley: University of California Press.
Farrell, J. B. (1977). Validity and rationality: The rhetorical constituents of argumentative form. Journal of the American Forensic Association 13, 142-149.
Faulkner, W. (1954). Accepting the Nobel Prize. In: H. Peterson (Ed.). A Treasury of the World’s Great Speeches. (Pp. 814-816), New York: Simon and Schuster.
Gross, A. G. And M. Dascal. The Question of Conceptual Unity of Aristotle’s Rhetoric. Colloquium, Wayne State University, Detroit, January, 1998.
Lincoln, A. (1954). House divided against itself cannot stand. In: H. Peterson (Ed). A Treasury of the World’s Great Speeches (pp. 491-97), New York: Simon & Schuster.
McBurney, J. (1936). The Place of the Enthymeme in Rhetorical Theory. Speech Monographs 3, 49-74.
McKerrow, R.E. (1977). Rhetorical validity: An analysis of three perspectives on the justification of rhetorical argument. Journal of the American Forensic Association 13, 133-141.
Perelman C. And L. Olbrechts-Tyteca. (1969). New Rhetoric: A Treatise on Argumentation. Notre Dame, Indiana: University of Notre Dame Press.
Toulmin, S. E. (1958). The Uses of Argument. Cambridge: Cambridge University Press.
Zarefsky, D. (1981). “Reasonableness” in public policy argument: Fields as institutions. In: G. Ziegelmueller and J. Rhodes (eds.), Dimensions of Argument (pp. 88-97). Annandale, VA: Speech Communication Association.