Peer disagreement & Assessor Relativism

   

Assessor Relativism: Disagreement & Deep Agreement By: Mahan Esmaeilzadeh

Assessor relativism is often criticized for its inability to account for disagreement. Under this view, the truth of a proposition is relative to the context of assessment, which means that two apparent dissenters have no common ground in which they can pursue a single truth. In this paper I will argue that assessor relativism can account for our intuitions regarding disagreement in subjective domains such as moral discourse and matters of taste. To argue as such I will first explain the motivation behind proposing assessor relativism as an alternative to contextualism. Then, I will explicate the problem with assessor relativism. Next, I will propose that we can use Plunkett and Sundell’s notion of metalinguistic negotiation to recapture the sense of disagreement even when truth is relativized to a context. Finally, I will raise a practical worry in relation to my proposed solution and respond to it accordingly. Contextualism, as a response to our intuitions about the subjective discourse, such as matters of taste and arguably morality, faces many problems, one of which is its inability to account for disagreement. Ragnar Francén identifies 4 distinct types of propositions in relation to our intuitions about disagreement in relation to them: 1) Propositions that are expressed by sentences containing indexicals. For example when I say ‘I am Mahan’ and you say ‘I am not Mahan’, we do not think that there is any disagreement whatsoever (Francén page 22). 2) Propositions about matters of taste. We do not treat a disagreement about the taste of chocolate as we treat the above “dispute” about my name. But at the same time we are reluctant to treat the disagreement over such discretionary beliefs akin to our treatment of objective matters of fact. Therefore, Francén argues that we have unstable intuitions

 

1  

    about this group. That is though at first it seems as if we disagree about some matter of fact (e.g. deliciousness property of chocolate), upon reflection we recognize that the truth of such propositions is relative to the speaker. In face of disagreement, there is little to no incentive to aim for convergence. We are content with “agreeing to disagree”. 3) Propositions of moral discourse. Francén argues that moral claims fall somewhere in between subjective taste and objective fact. Though we may not wish to adopt the moral realists view, when faced with disagreement, we treat our propositions as objective. In this sense, we “have stable intuitions that moral disputes are disagreements about some matter of fact” (Francén page 22). 4) Proposition about objective, non-evaluative, facts. The contextualist treats propositions of the second and third type like that of the first, that is they claim that the sentences expressing those propositions contain implied indexicals of the first type. For example, the proposition expressed by ‘Murder is wrong’ would be identical to the one expressed by ‘In my view murder is wrong’, that is both express . On the surface, this might seem desirable since it assigns the presumed proper truth-value to each individual’s moral claims, but it fails to account for disagreement in these two domains, especially in moral discourse wherein we have stable intuitions and are reluctant to revise our views. Contextualism cannot account for disagreement, because such pairs of propositions as and are consistent with one another, and it is necessary—though not sufficient—to have conflict in content for disagreement, that is the same proposition needs to be disputed (Plunkett and Sundell 2013; Binderup 2008).

 

2  

    The answer to this worry is assessor relativism according to which not only the proposition expressed (i.e. content) depends on the context in which it is asserted, the truth-value of that content is in turn relative to the context in which it is assessed (Schafer 2012; Kölbel 2004). This approach makes it possible for there to be two people who assert and deny the same proposition—in the same world—and yet neither is mistaken. Either assertion is true relative to a different context of evaluation. This is easy to digest when one looks at assertions about matters of taste. When I say ‘chocolate is tasty’ and you respond by ‘no, chocolate is not tasty’, though we disagree over the truth of the same content, both assertions are proper, for mine is true relative to my standards of taste (i.e. context of assessment) and yours is true relative to your standards of taste. In other words, we are faultlessly disagreeing. This idea of there being faultless disagreements is well argued for by Kölbel (Kölbel 2004). According to Kölbel, each person occupies a perspective, which functions as one parameter in the context of evaluation to which the truth of all propositions is relativized. The context of evaluation of any proposition consists of three parameters, the world in which it is assessed, the time at which it is assessed and the perspective of the person who judges its truthvalue (Brogaard page 400). This means that the same proposition, concerning the same world, can come out as both true and false, since truth is relative to the perspective of the assessor. Assessor relativism, then, fares better than standard contextualism, for it is in line with our intuitions that we disagree in moral discourse and matters of taste, and yet no one is at fault. The worry here is that relativism, contrary to what it claims, cannot account for our intuition that there is disagreement. More specifically, as soon as one relativizes the truth of a proposition to the perspective of each individual, given that there is no common perspective between dissenters, there is no disagreement. Binderup best articulates this worry. She writes:

 

3  

   

If the participants accept the perspectivalist semantics of moral discourse [i.e. assessor relativism], all rational incentive to engage in further debate would evaporate, leaving a general tolerance of moral disagreement. If it were true that circumstances of evaluation of moral utterances were relative to the (moral standards of the) speaker of the context of utterance, and if everyone knew this, then there would be little point in engaging in moral debate with others with different moral standards, since there would be no common single truth to aim at. The very point of moral debate would have disappeared, at least for semantically clearheaded users of moral language. (Italics mine; 412)

When I assert and you deny one and the same proposition , given my moral standards, it will come out as true, and given another’s moral standards, it will come out as false. Kölbel says that there is no need to worry, for this is simply an instance of a faultless disagreement. More precisely, there is conflict in content, and hence the disagreement, but since the conflicting propositions are not held in the same perspective, it is fault. Francén has a different take on the issue. She makes a distinction between contradictory assertions and disagreeing persons (page 29). Whereby Kölbel thinks that we have faultless disagreement, there is merely conflicting contents, that though necessary for disagreement, it is not sufficient. To see how, let’s look at the distinction between the two (Francén 29).

(Contradictory assertions) there is a circumstance of evaluation, C1, such that both assertions concern C1, and the two propositions asserted cannot both be true at the same circumstance of evaluation.

 

4  

    Simply put, two assertions express conflicting contents, if they concern, that is they pertain to, the same perspective but cannot both be true. Counterfactually a case where I assert P in world A and you assert not-P in world B, does not constitute a conflict in content, for the assertion is usercentric, and hence the contents concern distinct worlds.1

(Disagreeing persons) there is a circumstance of evaluation, C1, such that each person holds some proposition to be true at C1, and the two propositions cannot both be true at the same circumstance of evaluation.

If I assert P, and I take it that it is true of world A, and you assert not-P, and you too take your assertion to be true of world A, and our two assertions cannot both be true given a single perspective, then we are disagreeing. But in assessor relativism, it is not the assertion and denial of the same proposition relative to the same context of evaluation that is at play. Simply put, you and I may even agree on the truth-value of P relative to every context. Since, when we make our assertions, we don’t specify a specific context—and the default context is the speaker’s—you can acknowledge that P is true relative to my context and I can accept that its negation is true relative to your context. Therefore, even though we have conflict in content, we don’t have disagreement. Since there is no single truth of the matter to debate over for the “semantically clearheaded” users of language, any sense of genuine disagreement dissolves. As soon as we become aware of the semantics of moral discourse and matters of taste, we would find it odd to carry the conversation any further than a simple assertion of our beliefs. In fact, as long as one is internally consistent, all of one’s (everyone’s) assertions will be true simpliciter. I a priori know                                                                                                                 1

 

The world case is an analogy, for it is distinct from perspectives.

5  

    that whatever proposition of the second and third type I assert, I will always be correct, since the relevant context of assessment of my assertions will be my own standards (Binderup page 4123). Given this line of argument, it seems that to account for disagreement, it is not enough that the apparent dispute is over the same content as assessor relativism guarantees. The truthconditions of the disputed content must not vary between disputers. In effect, the objection is that though the assessor relativist can ensure conflict in content, in return it loses the disagreement that we intuitively feel in moral discourse and matters of taste. I wish to argue that there is a way for the assessor relativist to account for our intuitions about disagreement without having to give up relativization of truth to perspectives. To argue for my position I will make use of Plunkett and Sundell’s notion of Metalinguistic Negotiation (page 3). They use this notion to argue for the thesis that one does not need to share meaning in order to argue. That is not how I use it here. What I do take from them is the idea that by uttering a sentence, we express two distinct contents; that is we express its semantic content and via that we pragmatically convey the intended content. In sarcasm for example, one says (semantically expresses) one content (often a compliment) through which one pragmatically coveys the intended content (often an insult). Plunkett and Sundell make use of this distinction to show that sometimes by uttering a sentence in a conversation (e.g. an apparent disagreement), one is pragmatically advocating for the specific way they have used their expressions. Or to put it in technical terms, one metalinguistically uses an expression to lobby for their preferred semantic meaning of it. The crucial point is that the two contents are expressed and conveyed via the same sentence.

 

6  

    Plunkett and Sundell think we can metalinguistically advocate for a particular semantic meaning. I, on the other hand, think that we can do the same but at a deeper level. What if we could metalinguistically negotiate over something other than one’s preferred semantic meaning of an expression? What if we could metalinguistically advocate for our own perspective? Let us examine an example to assess the feasibility of this project. Imagine a Kantian and a Utilitarian debating over the moral permissibility of murder. K: ‘It is never permissible to murder’ U: ‘ It is not the case that it is never permissible to murder’ K and U semantically express the same content, albeit one affirms and the other denies it. Per the assessor relativist’s view, the truth-value of the expressed content is determined by K’s and U’s contexts of evaluation, and as such K’s assertion is true relative to his perspective, and U’s is true relative to hers. At this stage, we have conflict in content, but lack any genuine disagreement. And I want to say that that is fine, for the disagreement is not about the truth of the expressed content, but about which perspective to adopt. For example, if you and I disagree about whether or not P, and I know that you will accept P were you to believe Q first, I would try to convince you of Q first. This is a rough analogy but it does highlight the main point that in asserting that P, I am trying to change your mind about Q so that you will accept P as well. The pragmatically conveyed content of the sentences uttered by K and U is one that aims to change the other’s perspective in order to change his/her stance on the truth of the semantically expressed content. When K asserts ‘It is never permissible to murder’, by using ‘permissible’ as he does, he is metalinguistically advocating for his own Kantian perspective. So is the case with U.

 

7  

    I do not believe that this is an outlandish picture of what is in fact happening. We are often aware that other cultures value different things than we do. We know that it would not make sense to get them to agree with us right off the bat; instead, through expressing what we find valuable, we try to shift their moral perspective to one that is closer to ours. This, I believe, is even a deeper form of disagreement than what the assessor relativists presumed they could account for. Furthermore, I believe that my account provides the beginning of some theoretical support for Schafer’s Aim of Straightforward Conversation, according to which,

When one asserts an assessor-sensitive proposition in straightforward conversation, one aims to bring it about that this assertion has the same truthvalue relative to the context of assessment of every person in one’s conversational context (page 616-7).

In other words, we wish to bring about a convergence in the contexts of assessment of the parties to one’s conversation (page 617). Only through metalinguistic negotiation can one hope to bring about such a convergence of contexts of assessment. One last interesting consequence of my account is that, not only it gives us disagreement, it gives us true agreement; something that assessor relativism cannot account for. Something that is overlooked in the literature on assessor relativism is the fact that given that truth is relative to context of assessment, when two people agree that P, they may not agree for the same reasons. Of course for all intents and purposes they agree, but their agreement lacks intuitive depth. For example both a pediatrician and a pedophile agree that they should save a child, but their agreement is skin-deep. They affirm the same proposition but due to radically different perspectives. We (or at least I) have the intuition that there is no real agreement here. My

 

8  

    account, on the other hand, can account for our desire to converge on a deeper level than just the truth of one single proposition in isolation. By metalinguistically advocating for our respective perspectives, not only do we want to bring about agreement, we want to agree for the same reasons. The main question that remains is How? How does one advocate for a particular perspective and successfully change someone’s perspective? How can we even begin to change another’s perspective given that there is no objective fact of the world that we can point to in order to facilitate (if not necessitate) the transition from one perspective to another? The answer to this worry largely depends on the discourse in question. For example if it is another’s taste one wishes to influence, it is best to gradually introduce him to the cuisine, style of dining and the norms of the particular taste that one has in mind. No one is born with a taste for expensive wine for example; rather we learn to appreciate good wine over time. In other words we form wine-perspectives. When it comes to moral discourse, the answer need not be any more complicated than how one influences taste. What I have in mind runs along the same lines as Richard Rorty’s pragmatism in moral discourse. According to Rorty, we can influence others to adopt our “human rights culture” by manipulating their sentiments through stories (page 170). Rorty believes that there are no rational bases for accepting one set of moral standards over another, and that the only way to get others to adopt our standard is through emotional manipulation. I am not advocating Rorty’s view here, but merely offer one possible way of understanding how one can shift another’s moral perspective. These two examples make salient the point that though one metalinguistically advocates for a perspective, the actual convincing is done through a range of practical and theoretical

 

9  

    chores. It should not be surprising that, in most subjective domains, changing another’s perspective is no easy task. To convince a Kantian that murder is permissible, it is not enough to pragmatically advocate for that view through one single assertion that ‘murder is permissible’; much more work needs to be done. In conclusion, I believe that not only assessor relativism can account for deep disagreement; it offers a better picture of agreement as well. I proposed that it is possible to understand disagreement in moral discourse and matters of taste in terms of each side pragmatically advocating for their own perceptive in order to get the other party to agree with them. This modified version of assessor relativism can account for our intuitions about disagreement, and then some. Bibliography Binderup L. “Brogaard’s Moral Contextualism”. The Philosophical Quarterly, Vol. 58, No. 232 (Jul., 2008), pp. 410-415. Brogaard B. “Moral Contextualism and Moral Relativism”. The Philosophical Quarterly, Vol. 58, NO. 232. July 2008. pp. 385-409. Francén R. “No Deep Disagreement for New Relativist”. Philos Stud (2010) 151: 19-37. Kölbel M. “Faultless Disagreement”. Proceedings of the Aristotelian Society, New Series, Vol. 104, 2004. pp. 53-73. MacFarlane J. “Relativism and Disagreement”. August 17, 2006. Plunkett D., Sundell T. “Disagreement and the Semantics of Normative and Evaluative Terms”. Philosopher’s Imprint, Vol. 13, NO. 23, December 2013. Rorty R. “Truth and Progress”. Article: Human rights, Rationality and Sentimentality. Cambridge University Press. Philosophical Papers, Vol. 3. pp. 167-185. Schafer K. “Assessor Relativism and the Problem of Moral Disagreement”. The Southern Journal of Philosophy, Vol. 50, Issue 4, December 2012. pp. 602-20.

 

10