Relativism and internal reasons

 

 

The aim of this essay is to explore Bernard Williams' conception of relativism in ethics and its relevance for ethical disputes. More exactly, I am interested in pointing out what is the role of relativism in setting out the ethical disputes compared with the role of non-relativistic means deployed for the same task. Therefore, the relationship between Williams' station regarding relativism and his reduction of external reasons to internal ones is the core of this paper.

The strategy deployed is that of articulating his conception about the available means one has for handling ethical disputes. Due to this limited concern, I am not trying to find out gaps in his arguments, but on the contrary, I am taking the best case for his theses while concentrating upon the theoretical implications of them.

The reason why I am concerned with the relationship between the reduction of external reasons to internal ones and relativism is that both of them are seen as means to explain away ethical disagreements. In the first case, Williams' point is that external reasons could not be conceived as buttressing an ethical position, but internal reasons alone should be understood as relevant to ethics [see IER: 368-369]. Therefore, any ethical disagreement is a matter of deliberative reasoning around the subjective motivational set of an individual. This method is available when involved are mere individuals; however, the solution reached at the level of mere individuals' interaction is not subjective but an objective one because the solution is given in ethical terms that are shared by individuals' society [see MIE: 23; ELP: 162, TR: 139]. In the case when such a shared ethical language is not available, namely when one talks about disagreements between societies [MIE: 19, ELP: 162], relativism is the solution one could use in order to settle the ethical conflicts [see ELP: 156]. Hence, internal reasons are to be used when one talks about the ethical problems between mere individuals, while relativism is to be used when one talks about the ethical problems between societies. Nevertheless, a difficulty rises when Williams recognizes that Gilbert Harman's version of relativism respects the conceptual constrains required in order to meaningfully talk about relativism [see TR: 142, footnote 5]. This is so because Harman explicitly maintains that relativism is intelligible when one refers to disputes between mere individuals only [MRD: 190], and this relativism is possible exactly due to the some inner judgements that contain the motivational attitudes of individuals [MRD: 194-196]. Now, if Williams is right in accepting Harman's relativism as intelligible, and if Harman's is right in his reasons for relativism at the individuals' level, then Williams position seems to be self-defeating. Plainly stated, if this is the case, then Williams' understanding of relativism is inconsistent since the cases which are not recognized as falling under relativism (i.e., the internal reasons) turn out to be, in the end, cases of relativism.

Nevertheless, I am not interested here in verifying the soundness of Harman's relativism and its connection with Williams' internal reasons standpoint. But his suggestion is the reason why I think that one should take a closer look to the relation between internal reasons and relativism. Therefore, I present the details of Williams' station regarding the cases falling under relativism and the cases not falling under it. My concern will be neither the internal strength of internal reasons thesis as such nor the internal strength of the constraints required in order to meaningfully talk about relativism, but the relationship and coherence between these two points about ethics.

 In Morality. An Introduction to Ethics, Williams divides subjectivism intro three different kinds [MIE: 14]. The most primitive is the logical or linguistic subjectivism, which states that ethical judgements are nothing more than one's expression of one's attitudes. This kind of subjectivism is either not affecting the ethical disputes, or, if affecting them, is not providing grounds for ethical deliberation, and therefore one needs a different kind of subjectivism [MIE: 16-19].

A more refined form of subjectivism is the epistemological subjectivism, which maintains that the ethical statements can't be proved as true, but they are matters of opinion. At this point, the argument is split in two; namely, if one is confronted with epistemological subjectivism at the level of individuals, then one should perceive that "there are limits to the degree of ultimate disagreement that can exist within society - for without some degree of moral homogeneity it would not be a society" [MIE: 19]. Therefore, either subjectivism does not affect ethical deliberation or it needs a stronger formulation (the stronger formulation is the metaphysical subjectivism). On the other hand, if there one is confronted with the epistemic subjectivism at the level of societies, then the alternative for reaching an agreement is relativism [MIE: 20, ELP: 156-157]. However, one could not use  relativism in the cases where there is a real confrontation between societies [TR: 138], which is equivalent with either the possibility of using the same language of appraisal in their dispute [ELP: 161], or that one society is the mere forerunner of the other one [ELP: 166-167], or that for those implied in dispute is possible to move to either of the conflicting societies [TR: 139-140].  As it will be refined a little bit later, relativism is available for solving out ethical disputes in the cases of notional confrontations only.

If relativism is not at hand, then the subjectivist may move to the vaguest kind of subjectivism, that is, metaphysical subjectivism. Metaphysical subjectivism holds that there are no moral facts [MIE: 14], and the claim is that even "if two observers are in the same observational situation, have the same concepts, etc., then they are [not] bound to reach the same moral beliefs" [MIE: 31, author's italics]. However, this kind of subjectivism is based on a mistaken analogy between what is happening in the field of science and what is happening into ethics. Precisely, the defender of metaphysical subjectivism says that if it is rational for one to suspend judgement if one could not solve a conflict between two scientific claims, then it is rational for one to suspend judgement if one could not solve a conflict between two ethical claims, too. But this analogy is misleading because "the vital difference is that the disagreement in morality involves what should be done. Once you see this difference, you see equally that it could not possibly be a requirement of rationality that you should stop caring about these things because someone else disagrees with you" [MIE: 34]. To hold this kind of subjectivism is equivalent with being inconsistent, like saying that "I am convinced that racial discrimination is intrinsically wrong, but it is possible for all that, that it is not" [MIE: 35].

Therefore, subjectivism is theoretically acceptable only as an epistemological claim, and only at the level of interactions between societies. Nevertheless, this is not a wide-spread conviction; e.g., for Ayer, non-cognitivism in ethics does not imply relativism [see Ayer, A. J. "On the Analysis of Moral Judgements." Philosophical Essays. London: Macmillan, 1954, p. 247]. This is so because his positivist starting point that all the synthetic sentences are empirical. The strategy employed for sustaining the claim that all the synthetic propositions are empirical is that of showing that the ethical propositions are neither empirical nor synthetical; indeed, they are not intelligible as propositions, because they lack any criterion for being verified. The ethical assertions are not verifiable because they are not descriptive; they are not descriptive because the ethical terms are pseudo-concepts, which add nothing to a factual description, and therefore they are unanalysable [Ayer, A. J. Language, Truth, and Logic. London: Penguin Books, 1990, chapter 6]. That is why an epistemic relativism in ethics could not be started from a non-cognitivist viewpoint. But for Williams subjectivism does imply relativism, and a subjectivism that rejects relativism is weaken [ST: 204-207]; indeed, the merit of subjectivism is that it offers the possibility of using relativism as a means to solve out conflicts which are, otherwise, unsolvable.

Aside the sphere of relativism and those inter-societies cases in which it can be applied, the ethical conflicts are to be solved by deliberation. Nevertheless, even if subjectivism has been circumscribed to some special cases, it does not mean that ethical disputes may be objectively solved, that is, in terms of truthfulness or falseness. They may be settled out by a special understanding of ethical objectivity - namely by grounding them in considerations about human nature, considerations arrived upon by some deliberative convergence of practical reason [see ELP: 152-154]. These considerations about human nature use moral obligation as one of their means for solving out disputes (moral obligation is a special case of deliberative conclusion) [ELM: 174]. Nonetheless, the language of appraisal is used to the same purpose [EML: 161, 193; TR: 141]. My concern in this point is with the role played by the internal reasons; for this, I first present the basic points which divides the sphere of relativism from the rest of ethical disputes.

In order to be able to use relativism for explaining away an ethical conflict [ELP: 156], one should first be able to talk meaningfully about relativism. The logical pre-conditions that set up the theoretical framework for relativism are the following.

(1) The first logical condition for relativism is to have  a set of at least two different social systems of belief (S1 and S2). This is a mere definitional condition, i.e., it is impossible to talk about relativism if there is a set with only one social system of belief to choose from [TR: 132].

(1.1.) Therefore, the members of the S' set should be self-contained, which means that every social system of belief from S' set is recognized as a social social system by those who share it [TR: 133]. For instance, if S1 is the mere forerunner for S2, then those who actually share S2 may not be able to distinguish between S1 and S2; in this case, one would not be able to meaningfully talk about relativism [ELP: 164-165].

(1.2.) Due to the difficulties of deciding if an S is self-contained or not when compared to other S (as the above example indicates), a further refinement is the condition that those social systems of belief should be exclusive to each other; i.e., the Ss from the given set should have their separate identities secured [TR: 134]. Notwithstanding, the exclusiveness of two social systems of belief does not necessarily imply that one will be able to meaningfully talk about relativism. That is, "a fully individuable culture is at best a rare thing. Cultures, subcultures constantly meet one another. Social practices could never come forward with a certificate saying that they belonged to a genuinely different culture" [ELP: 158]. For this reason, the criteria one should use in order o ensure that the set of Ss is composed from different Ss needs further refinements.

(1.3.) The refined logical pre-condition for a meaningful conceptualization of relativism is that the Ss should have conflicting consequences in order to secure that there is a set of different social systems of belief [TR: 134]. For fulfilling this condition, two additional points are required.

(1.4.) In order to be able to talk about Ss' conflicting consequences, one needs the comparability condition for these consequences. Precisely, there should be at least one possible description of Ss' consequences in acceptable terms for all the social systems of belief. If this requirement is not fulfilled, then one cannot be sure that the consequences are really conflicting, which means that one cannot securely talk about different social systems of belief, and therefore one cannot meaningfully talk about relativism. [TR: 134] But all to be said would be that the Ss are incommensurable exclusive [TR: 137], which means that the conflict between Ss is not solvable (one would have to say that 'it is impossible to live within both Ss', period. [TR: 135])

(1.5.) However, it could be the case that the condition 1.4. is fulfilled, but one sees that in the light of a common description, one of the alternatives does not hold anymore (e.g., it is internally incoherent). This is why it is needed the requisite that if one gives a clear-cutting yes/no answer to the question of choosing one of these comparable but conflicting Ss, then all the Ss will still hold [TR: 134-135], that is, there will still be a set of alternative Ss to choose from.

These conditions required in order to be able to meaningfully talk about relativism may be satisfied both diachronic or synchronic [TR: 137]. In both cases, if the social systems of belief are in mutual ignorance, then the comparability condition is unsatisfiable; therefore, relativism may not be used as a means for solving out disagreements between them [TR: 138]. Nevertheless, if they are ignorant towards each other, then not even relativism is difficult to be conceptualized between them, but disagreement also seems hardly conceivable.

If the social systems of belief are not in mutual ignorance and the above conditions are satisfied, then the disagreement between Ss may be solved through relativism only when other means (objective in the special sense stated earlier) are not available. If these other means are not available, then the confrontation is mere notional and one can use relativism in trying to solve it out. Whenever objective means are at hand, the confrontation is understood as a real confrontation and is settled through them (relativism is excluded). A real confrontation between social systems of belief means

a) either that these systems of belief are real options for those involved in disagreement [TR: 138](synchronic Ss),

b) or that one social system of belief is the mere forerunner of the other one [ELP: 166-167](diachronic Ss),

c) or that there is the possibility of using the same language of appraisal in dispute [ELP: 161](both diachronic and synchronic Ss),

d) or that the confrontation is between mere individuals pertaining to different social systems of belief [ELP: 162, MIE: 19, 23](both diachronic and synchronic Ss).

 

a) An S is a real option for an individual or a community X if either (i) X already holds S, or (ii) X' transition to S is possible. Condition (ii) means both that X "retains its hold on reality", and that X' transition to S is conscientious and rational. More exactly, the question if an S is a real option is an objective question (or/and a social question), because the possibility of transition to an S2 depends upon the objective feature of the original S1, and the changes needed for accepting S2. This simply means that X' subjective opinion about S2 does not count in establishing if S2 is a real option or not. X' opinion is not sufficient for deciding about S2 because X may be "ill-informed, unimaginative, un-self-aware, or optimistic". On the other hand, X' opinion is not necessary for deciding about S2 because S2 may have unknown features for X, which can make S2 a real option regardless X’ contrary opinion [TR: 140].

Due to (i) and (ii) of this case, solving the disagreement between Ss cannot be done through relativism due to the fact that condition 1.5. would be infringed; that is, due to the fact that involved are real options, one of the Ss will be objectively reduced to the other, which means that there is no set of different Ss to choose from, and relativism would be unintelligible.

b) If S1 is the forerunner of S2, then the dispute between S1 and S2 is  not falling under relativism simply due to the fact that S1 and S2 are not clearly distinguished.

c) If one is able to substantively use the appraisal terms about different  Ss, then one recognizes something that links the Ss; otherwise the language of appraisal would be used in a quite unproper manner. For Williams, if one is capable of a proper use of vocabulary of appraisal, then this is a sign that the Ss to which this vocabulary is applied are real options for one [TR: 141]. Consequently, the disagreement between Ss will be solved out through reasons relevant for individual, which are nothing more that the internal reasons.

d) If in the Ss dispute at stake are the practical decisions that some particular individuals should take towards each other, then it will be a mistake to see the dispute as a subjectivist will see it (see the treatment of metaphysical subjectivism above). As Williams puts it, "it is essential to morality and its role in any society that certain sorts of reactions and motivations should be strongly internalized, and these cannot merely evaporate because one is confronted with human beings in another society" [MIE: 23]. Even if there are no common values between conflicting Ss nor are the Ss real options for one, still the conflict should be solved in a straightforward manner. I would say that this case is falling under internal reasons solution, too. This is so because an individual would have to see which is her moral obligation towards another individual; nevertheless, this is exclusively done through internal reasoning, as we will see.

 

If the confrontation between Ss is not a real one, then it is a mere notional confrontation. Moreover, if the logical pre-conditions for relativism are fulfilled, then one may use relativism as a means to solve out the disagreement between Ss. Now, the fact that all these conditions are fulfilled does not imply that relativism is necessarily at hand; there are some kinds of relativism that one should reject due to some internal contradictions or explanatory insufficiency. That is why one should reject relativism based upon confusion or misunderstanding. For instance, if X confounds X's actual S1 with S2, then one is not entitled to describe the confrontation between S1 and S2 as a matter of relativism because this would infringe relativism's requirements, but as a real confrontation [TR: 143].

The largest kind of avoided relativism is the relational relativism [ELP: 156]. The relational relativism uses such phrases as 'true for us' or 'true for them' [TR: 142]. The first kind of problems for relational relativism is related to the above-mentioned logical pre-conditions for a meaningful notion of relativism. On one hand, this relativism, based upon the relation 'true for us' vs. 'true for them', infringes the comparability condition 1.4.; this relation ('true for us' vs. 'true for them') states nothing more than the incommensurability of positions [TR: 133], which is not a sufficient condition for stating the fact that their consequences are conflicting [ELP: 157]. On the other hand, there is no such a thing as a fully individualized social system that would allow one to use the relation 'true for us' vs. 'true for them' in order to secure the needed logical pre-conditions for a substantial relativism [ELP: 158]. The second kind of problems for relational relativism is that it may not suitably perform the explanatory task for which was deployed. Plainly, if relativism is to be used as a means to solve away a conflict, then it should perform two tasks; namely, it has to show why there is not conflict and why it looked as if there were one. While relational relativism seems successful in what regards the first task, "it tends to have less success with the second task because, the more convincing it is to claim that the statements are really relational, the more puzzling it is that people have thought there was a conflict" [ELP: 157]. For example, the ethical relativism of Gilbert Harman is a relational relativism; and it is due to this second kind of problems why it should be rejected, despite the fact that it respects the logical pre-conditions for relativism (as Williams himself acknowledges) [see footnote 1, ELP: 219 and footnote 5, TR: 142].

Other kind of refuted relativism is the vulgar one [MIE: 20-25, TR: 143, ELP: 159]. Vulgar relativism is a sub-species of relational relativism; besides the difficulties of relational relativism, vulgar relativism is also loaded with a supplementary internal contradiction. That is, it uses at the same time a concept in both relative and non-relative sense [MIE: 20]. To put this into another form, vulgar relativism treats real confrontations like notional ones [TR: 143]. All in one, it would be a mistake to designate a logical error as relativism.

Despite the above-stated case c) that specifies when one should take a dispute between Ss as a real confrontation, it could be the case that the involved Ss are not real options for one, i.e. there is no real confrontation between Ss but still one may use appraisal. This special case of notional confrontation is named relativism of distance; the distance of one of the Ss ensures that that S is not conceivable as a real option for those involved: "distance makes confrontation notional" [ELP: 162]. Here, distance should not be understood as merely spatial distance, but as conceptual distance; specifically, the relativism of distance is to be applied to exotic, past and future Ss. In any case, the fact that it is used the appraisal language means reference to internal reasons.

The internal reasons for an action could be formalized as ‘individual X has a reason to do Z, where Z will serve X’s motive’; that is, the internal reasoning is linking the action Z with an internal motive which X already has, where Z is discovered through this reasoning [IER: 363]. Therefore, internal reasoning tries to find means for satisfying X’s internal motives. The internal motives of an agent are named the agent’s subjective motivational set (S). The elements of subjective motivational set are indeterminate [IER: 369], but one could say that between them are desires, dispositions of evaluation, patterns of emotional reaction, personal loyalties, various projects and commitments of agents. Mere needs, as needs, are not to be conceived as elements of one's subjective motivational set simply because a need still requires a motif for being followed [IER: 366].

The internal reasons starts from a very simple assumption, namely that if there is an element of one's subjective motivational set which may be satisfied by a certain action, then one has a sound reason to perform that action. Due to the fact that this starting assumption ignores many problems related to the elements of one's subjective motivational set, this basic model is named a sub-Humean model. The refinements of the sub-Humean model are as follows:

            i. “An internal reason statement is falsified by the absence of some appropriate element from S” [IER: 364]. This statement implies for a sub-Humean model that all the members of S may give rise to internal reasons. For a more than sub-Humean model, this statement implies that one should exclude the false beliefs from S (because internal reason concerns all the agent’s rationality, not only moral explanation). Therefore, a further refinement of the sub-Humean model would require that

            ii. “A member M of S will not give to an agent a reason for Z-ing if either M’s existence is dependent on false belief, or agent’s belief in the relevance of Z-ing to M’s satisfaction is false.” This would mean that

            iii. When an agent have a false internal reason about herself or when she is ignorant about an internal reason about herself, one could not properly say that in these specific cases the agent should or should not perform a certain action (speaking properly, the reason is missing).

            iv. In order to eliminate the above difficulty, one needs a criteria for finding out correct reasons. The criteria used for internal reason statements is that they can be discovered in deliberative reasoning starting from S’ elements. As a result, the practical deliberation in Williams’ internal reasoning model (which is not the sub-Humean model anymore) has three functions. First, it points out new actions that an agent should do, starting from agent’s S. Second, it could give additional reasons for an action Z which is already motivated. Last, it influences S’ content by uncovering the false elements in S. [IER: 365-366]

            The competing viewpoint for internal reasons is the claim that reasons alone, by demonstrating their propositional truth, may motivate a rational agent to perform (or not) an ethical act without any appeal to individual' subjective motivational set. For Williams, this claim is either reducible to the claim that internal reasons are those that matter in ethics, or impossible to use.

            On the first hand, in order that an external reason to motivate an agent, the agent should believe that an external reason could motivate him. However, believe is subjective and not related to reason. Therefore, the external reason could be either a link between an element of S and an action, but in this case the external reason is identical with the internal one; or it could give rise to a new motive in order to act, which is either self-contradictory (because external reasons claim excludes motivation) or it must involve a belief (which is part of internal reasoning) - therefore is confounded again with the internal reasoning. As a conclusion, in all the relevant and consistent cases for ethics, external reasons are reducible to internal reasons.

On the other hand, the external reasons theory could not affirm that an agent is irrational when the agent is not convinced by an external reason, simply because the agent could claim that he is not convinced and not performing a certain action because it may be the case that the external reason which is supposed to convince him is false, and therefore it would be irrational for him to let himself to be convinced. Moreover, Williams says that there would be no difference between false and truth external reasons if it is to convince one [IER: 369]; therefore the external reasons standpoint is of no use in ethics.

As I said at the beginning, I am not interested in checking out the soundness of Williams cases for either relativism or internal reasons; on the contrary, I take them in their best lights. My focus is to see what is the role of each of them in ethics and their relationship.

Until now, I have shown that relativism is a means for explaining out ethical disputes when there is a notional confrontation between clearly distinct  and stand-alone social systems of belief. Internal reasons are means for explaining out ethical disputes when there is a confrontation between individuals which share some elements of their subjective motivational set. The problem is the use of internal reasons both in the real confrontation between individuals pertaining to different social systems of belief (see d. above) and in the notional confrontation (the case of relativism of distance).

The point is that, due to its usage, it seems that internal reasons may be used both as an objective and a relative criteria for solving out ethical conflicts. The use of internal reasons ensure that the disagreement is objectively solved because the ethical stations, which ultimately depends upon the subjective motivational set of the individuals, are not random decisions of the individuals, but they rest upon individuals' confidence in their ethical stations. As this confidence is a social phenomenon, then ethical stations are quite objective (remember the revised sense of objective)[ELP: 169-171; 192-193] Another way to say this is the one I already pointed out, namely that the subjective motivational set is subject to the influences of deliberative reasoning.

However, this is not clearly happening in all the cases when internal reasons are supposed to be used. The first case is that of a real confrontation between individuals whose subjective motivational set have been influenced by different social practices and deliberations (see d). If the case d. is objectively solvable, it is not clear at all how this task is to be performed. If the case d. is not objectively solvable, it means that all the arguments for relativism which are based upon the assumption that relativism may take place only between social systems, even if not vanished, should be revised.

The second case when it is not clear how internal reasons may help is their use for solving an ethical dispute in the case of the relativism of distance. If internal reasons are helpful, then the relativism of distance is vanished, because the use of internal reasons implies that the case of relativism is a trick, an objective solution using the mask of epistemological subjectivism. If internal reasons are not helpful, then relativism of distance may not be used for solving out ethical disputes. On the contrary, one would need Harman's relativism at the level of individuals if one is to solve a dispute using both relativism and the internal reasons theory, which whould be paradoxically.

I would say that the role of internal reasons is not clearly defined as regarding Williams' claims about relativism. One solution could be to understand the talk about relativism as a second order ethical issue, and the internal reasons as the concern of first order ethics. In this case, there would be no much trouble regarding the relationship between relativism and internal reasons. However, Williams says that Ayer is exagerating when denying the implication of the first order ethics upon second order ethics and vice versa [ST: 200], which means that he is commited to the view that between second order and first order the relationship should be explained. Therefore, the connection between relativism and internal reasons should be stated more clearly.

 

Cristian Vasilescu

2000

 

 

Bibliography

 

Bernard Williams, The Truth in Relativism, chapter in "Moral Luck", Cambridge: Cambridge UP, 1981, pp. 132-143 [Referred here as TR]

Bernard Williams, "Subjectivism: First Thoughts", "Interlude: Relativism", and "Subjectivism: Further Thoughts", chapters in "Morality. An Introduction to Ethics", Cambridge: Cambridge UP, 1993, pp. 14-25 [Referred here as MIE]

Bernard Williams, Subjectivism and Toleration, in  "A. J. Ayer Memorial Essays", A. Phillips Griffiths ed., Cambridge: Cambridge University Press, 1991, pp. 197-208 [Referred here as ST]

Bernard Williams, Internal and External Reasons, in "Moral Discourse and Practice", Stephen Darwall, Allan Gibbard and Peter Railton eds., Oxford: Oxford UP, 1997, pp. 363-371 [Referred here as IER]

Bernard Williams, "Relativism and Reflection", and "Foundations: Practical Reason", chapters in "Ethics and the Limits of Philosophy", Cambridge: Harvard UP, 1985, pp. 54-71, 156-174 [Referred here as ELP]

Gilbert Hartman, Moral Relativism Defended, in "Relativism. Cognitive and Moral", Jack W. Meiland and Michael Krausz eds., London: University of Notre Dame Press, 1982, pp. 189-204 [Referred here as MRD]