Two Methods

My current sense is that analytic philosophy has been burdened with the metaphysico-methodological assumption that reality is to be known in terms of univocal concepts and therefore the proper aim and method of philosophy is to isolate those singular meanings. Thus for example: What is a desire? What is morality? What is rationality? and so on. There is presumed to be one and only one correct or true answer in each case. This is the road not only to clarity but to truth in general, since the only way to know if a proposition is true (for example: “Morality is objective”; “Pluto is a planet”; etc. ad inf.) is to have a firm grasp of the meaning of each of its component terms (“morality,” “objective,” “planet,” “is”).

            Well, I’m with President Clinton on this one: It depends on what “is” (and “planet” etc.) is.[1] I now view concepts as polysemous, and so one does not really know what a term means unless that is spelled out as a disjunction, much as a dictionary provides several options in defining a word. Now, the univocal method also presumes several possible uses of a term, but the difference from the polysemous method is that they are assumed to represent either homonyms or, in all but one case at best, misuses of the term. The method I now employ presumes instead that, in addition to there of course being homonyms in some cases, as well as simple misunderstandings in others, in just about any and every case a term in a natural language has multiple meanings.

            So for example the typical analyst would insist that unicorns do not exist because there are no horses having a horn on their forehead. But an analyst like me would be relaxed about the question and grant that unicorns in that sense (probably or maybe) do not exist, but that unicorns in the sense of an animal with a horn on their forehead surely do exist. Nor do I think “unicorn” is simply homonymic in this example. People who are disagreeing about the existence of unicorns, it seems to me, intend to be talking about the same thing. (Just as two people disagreeing about whether a credit union is a bank are not confusing a financial institution with land beside a river.)

            I see two great advantages of the polysemic view: One is that it puts us on constant guard against ambiguity and equivocation and misunderstanding or miscommunication. This can help prevent or resolve countless unnecessary conflicts. The other advantage is that it undercuts an attitude of contempt for others whenever there is a disagreement based on the assumption that the other person must be mistaken. So, I predict, both the number and the intensity of unnecessary or exaggerated conflicts will be reduced by adoption of the method I propose. A third advantage, though I have hesitated to use that word for it, is that this method represents what is simply true about the world, or at least about our knowledge of it, namely, that it is typically (and perhaps of necessity) diverse. The necessity might be due to the dependence of meaning (and hence truth) on needs and purposes and goals (or “desires” as I tend to call them).

            An excellent example of the univocal approach is the (very untypical because extraordinary) book Three Faces of Desire by Timothy Schroeder. (See my review.) But, tellingly, Schroeder himself acknowledges that he is simply ignoring another kind of desire in offering his theory of desire. Contrast that to Chapter 1 (“Distinctions of Desire”) of my book It’s Just a Feeling, in which I simply lay out multiple ways the word “desire” is commonly used.[2] I could even see Schroeder as being at peace with my methodology (and me with his), claiming only to make the analysis of one of the types of desire more detailed or precise. After all, neither of us is doing mere lexicography: We want not only to “define” but also to draw out as many implications as we can for as many substantive issues as we can, namely, those that incorporate some notion of desire in their claims and counterclaims. Our chief interest may even be in those other issues. For me, for example, an issue of great moment is whether an ethics of desire can profitably replace morality (where of course I must also explain which use of “morality” I intend).[3]

            Another way to put (or I would put) the difference of methods is that the typical analyst seeks to justify claims (whether about the meaning of a term or about propositions containing that term) whereas I seek mutual explanations of different uses with regard to a particular issue and then leave up to my interlocutor or audience (or me for that matter) to find one or the other use more fitting and hence determinative of a resolution of the issue.[4]

Mitchell Silver (personal communication)  asks why the “new” method is not simply the old method, with each of multiple meanings being correct in a particular context. Thus (as I have done myself, e.g. in “Method and Meanings” in Chapter 1 of Reason and Ethics) distinct senses of a word could be labeled with subscripts, e.g., “morality_m” and “morality_e”; in effect they are different words, so this is a kind of homonymy after all. Paul E. Griffiths (in his excellent book What Emotions Really Are) takes a different tack with “emotion,” arguing that the general concept “needs to be replaced by at least two more specific concepts” (p. 247).

My response is (as always) yes and no: It depends (on one’s desires, purpose, etc.). The bottom line for me is that disputes about meaning or truth frequently arise from a failure to appreciate ambiguity, the realization of which would resolve or ameliorate the dispute. Whether the problem is to be understood as an unawareness of polysemy or of homonymy[5] can itself be a matter of preference. In other words, the method I suggest applies even to analyzing that method: There is more than one way to conceive it. None is “correct” tout court. I, naturally, have my preference due to various contingencies (causes), both known and unknown to me, and some of which I fashion as reasons in a rhetorical attempt (such as this essay) to make them persuasive to others.[6]

An example[7] of how this plays out: Suppose two people are arguing about the morality of abortion. The Pro-Choice (PC) person, who is also a vegan because she loves animals, accuses the Pro-Life (PL) person, who is an omnivore, of hypocrisy, since he inveighs against the killing of “babies” and yet feels no qualms about eating food that involves the killing of babies. The PLer objects that the animals she eats are not babies; babies are human. The PCer insists that babies are any young animals, human or otherwise (and most animals that we eat are young), whereas what is aborted is a fetus, not a baby.

My point is, first, that all of these meanings can be found in the dictionary. Therefore, second, everyone in this dispute is correct (to that extent). Third, nevertheless there is rhetorical advantage to be had by insisting on one’s own preferred definition (of “baby”), at least in the context of this debate. However (fourth) this also reinforces the intractability of the disagreement, for the simple reason that no one can ever definitively “win” a debate in which both parties are correct! It also (fifth) reinforces irrational stubbornness, since one is staking more and more on a definition rather than on substance.

So how does this bear on my methodological preference? Polysemists and homonymists agree that the logjam could be broken by recognizing that both uses of “baby” are legitimate: one refers to a human fetus and the other refers to any young animal. For the homonymist “baby” and “baby” are therefore simply two different words (like [river] “bank” and [savings] “bank”); for the polysemist there is one word having two different senses (like [savings] “bank” and [commercial] “bank”). My argument for preferring the latter is that in cases where people are disputing a substantive issue and relying on the same token (“baby”), it is easier to get them to acknowledge that the same word can have different meanings (or “senses”) than that they are using two different words. They have too much invested in the token. This is not the case for river banks and savings banks.

In either case, however, my empirical hunch is that once the verbal matter is clarified or, as it were, subtracted from the controversy, there will be more room for an engagement with the substance of the dispute because it can now be framed more intelligibly, to wit: Does a human fetus have a right to come to term that outweighs a woman’s right to decide whether she wants to have a child?[8] The business about young animals was a distraction (attributable to rhetorical overreach by the PLer and exploited by the PCer).[9] And just putting the issue in these terms can lead to both a speedier resolution and less mutual animosity (or so is my empirical hunch and wish).

The practical advice I wish to offer, therefore, is to clarify one’s meaning every step of the way when speaking or writing. (My own verbal expression has become peppered with “in the sense that.”) Hand in hand with this, to engage in any communication or intended communication (i.e., writing for a future readership), or any exchange, with the assumption that the other party will probably misunderstand you, and you them, without continual effort to clarify or seek clarification (by both parties).

Granted, this style of conversation can grate (or itself be misunderstood!) on those who do not understand the need for it, so, it can be hoped, one will develop techniques for making the medicine easier to swallow; for example, by withholding conversation whenever possible to occasions when there need be no rushing, resisting the temptation to introduce one’s queries by interrupting, and containing one’s own remarks in their pithiest formulations (albeit peppered with “in the sense that”).

In this essay I have diagnosed the underlying problem to be a kind of word worship that holds holy any word as univocal. Instead of this literalism, I have proposed an attitude towards words as conveyers of multiple meanings (understood as either polysemic or homonymic, though this is of secondary importance), whose precise meaning on any given occasion typically calls for inquiry.

---

[1] Contending his statement that "there's nothing going on between us" had been truthful because he had no ongoing relationship with Lewinsky at the time he was questioned, Clinton said, "It depends on what the meaning of the word 'is' is. If the—if he—if 'is' means is and never has been, that is not—that is one thing. If it means there is none, that was a completely true statement."  (Wikipedia)

[2] You could also check out my blog post, “What Is It to be Rational?” for a similar treatment of rationality.)

[3] Similarly I point out the use of “humane” in the regulatory literature for animal experimentation, as diametrically opposite its use in everyday discourse – my purpose and concern being the way other animals are treated in biomedical research (as well as in animal agriculture).

[4] When there is remaining irresolution, which may be the norm, ideally there will at least be a noncoercedly agreed-on peaceful procedure for deciding the question if there is a need to do so, for example, by taking a vote.

[5] Note that homonymy would then come in two degrees: river bank versus savings bank (different words), and savings bank versus commercial bank (different senses of the same word).

[6] Or I could be indifferent to whether others accept my interpretation of the method. (Or even wish they wouldn’t!) It depends.

[7] Adapted from Chapter 1 of my book Reason and Ethics.

[8] Even this phrasing does not capture the issue as I would prefer to articulate it, since it appears to presume the existence of moral rights, which I consider a fiction. So I think the real question is more like: “Do we who have the power to decide the matter want to live in a society that accords human fetuses legal priority (in the matter of whether to bring them to term) over the women who carry them?” In other words, this is about power and desires, not right and wrong. And I think that contemplating the realities and implications in toto of such a society would give most people pause.

[9] Again I can point to the use of “humane” in animal experimentation (and animal agriculture) as another example. It is rhetorical overreach for the experimenter to characterize her work as “humane,” and I as an anti-vivisectionist can exploit that. But the real issue is whether we humans want to continue to subject other animals to confinement, pain and suffering, and short lives on behalf of human well-being or pure research.