如何精读《斯坦福哲学百科全书》的词条（三）？吕炳强解剖Relativism

精读系列见：

第177期如何精读《斯坦福哲学百科全书》的词条（二）？吕炳强解剖Relativism

第176期 如何精读《斯坦福哲学百科全书》的词条（一）？吕炳强解剖Relativism

第83期 如何【精读细剖】ASR的理论论文？再来一次示范！（下）

第82期 如何【精读细剖】ASR的理论论文？再来一次示范！（上）

第62期如何【精读细剖】一篇AJS理论文章？来次示范吧！（下部）

第61期如何【精读细剖】一篇AJS理论文章？来次示范吧！（上部）

Text: 4.4.4Social Constructionism

[...]

4.5 Moral Relativism

[...]

5. New Relativism

Thereis a recent version of relativism according to which some of the viewsconsidered so far—for instance, Harman’s (1975) variety of moralrelativism—will be regarded varieties of contextualism as opposed tobona fide relativism. This recent version—sufficiently distinct from therelativisms so far considered that it is deserving of attention in its ownright—we are calling “New Relativism”, a variety of relativism that has arisenout of work in the philosophy of language in the analytic tradition, [...] Inthis section we aim to (i) outline several features that individuate NewRelativism; (ii) consider in turn motivations for (and objections to) severalprominent strands of it; and, finally, (iii) conclude with some philosophicalproblems that face New Relativism more generally.

5.1 The individuatingfeatures of New Relativism

Itis a commonplace that the truth-value of an utterance can depend on thecontext in which it is uttered. If you say “I’m happy” and I say the samesentence, your utterance may be true and mine false. In such cases, the contextof utterance plays a role in determining which proposition the sentenceexpresses. This can happen even when the sentence does not contain anovertly indexicalexpression. Thus Harman and Dreier hold that a statement of the form “Ais wrong” is roughly equivalent to “A is wrong according to the moral system Iaccept”. So twoutterances of (say) “Torture is wrong” can differ in truth-value if they areuttered by speakers who accept very different moral systems. Contextualistsabout (for instance) moral, aesthetic and epistemic discourse will viewmoral, aesthetic and epistemic expressions likewise as indexical expressionsbut (as we’ll see) with some difficulty explaining apparent genuinedisagreement in these areas of discourse. On this point, NewRelativists claim an important advantage over contextualists. Newrelativism, by contrast with contextualism, aims to achieve this advantage viaa much less familiar form of context dependence.

Truth-relativismwith respect to utterances in area of discourse Dis the claim that, following MacFarlane’s notable version of the view: the truthof S’s D-utterance u depends (in part) on a context of assessment; that is (andin short) what S asserts, u, gets a truth value—according to thetruth-relativist’s D-semantics—only once the D-standard of the assessor isspecified. Independent of the specification of such a standard,S’s u assertion lacks a truth-value much as, by comparison, indexicalexpressions such as “The barn is nearby” do not get a truth-value independentof contextualfacts about the context of use (i.e. the context in which the utteranceis made). And, as a further point of clarification here: while the contextualistcan, no less than the relativist, recognize a “standards” or “judge” parameter,for the contextualist, its value will be supplied by the contextof use, whereas the relativist takes it to be supplied [...] bythe contextof evaluation (or, as MacFarlane calls it, the context of assessment).

Tosee how this view is claimed to offer a satisfying take on disagreement,consider a simple example, concerning predicates of personal taste. A utters,“Pretzels are tasty”, and B utters, “Pretzels are not tasty”. While the semanticinvariantist (for whom the truth-value of taste predications is in no waycontext sensitive) will insist that the above exchange constitutes a genuinedisagreement about whether pretzels are tasty and that at least one party iswrong, contextualists and truth-relativists have the prima facie advantageousresources to avoid the result that at least one party to the apparentdisagreement has made a mistake.

Thecontextualistclaims that the truth-evaluable content expressed by A’s utterance encodes A’sstandards (cf. non-indexical contextualism). Thus, in this apparent disagreement,the proposition expressed by A is “Pretzels are tasty relative to my [A’s]standards” while B expresses the proposition “Pretzels are not tasty relativeto my [B’s] standards”. This maneuver avoids the result that at least oneof the two parties has uttered something false, but (as the newrelativist points out) this result comes at the price of being unable to offera clear explanation of our intuition that there is some uniform content aboutwhich A and B disagree.

Thenewrelativist, on the other hand, claims to be able to preserve both theapparent subjectivity of taste discourse and (and, unlike the contextualist)our intuition that exchanges of the form mentioned constitute genuinedisagreements. They do this by first insisting (unlike the contextualist)that there is a single truth-evaluable proposition which A affirms and B denies.In the case where A says “Pretzels are tasty”, and B denies this, there is auniform content that is affirmed by A’s utterance and denied by B’s, namely theproposition that pretzels are tasty, period. So we have a genuine disagreement.Unlike the truth-absolutist,however, the new relativist will add that the disagreement is faultless becausethe proposition affirmed in A’s utterance has a truth value only relative to a judge orstandards parameter, and in this case: A’s standards, when A is theassessor, B’s standards, when B is the assessor. Hence, the truth-relativistabout predicates of personal taste will, by insisting that the truth ofPretzels are tasty depends on the context of assessment, allow a singleproposition to be (at the same time):

(i)   true relative to the context of assessmentwhere A’s standards of taste are operative and

(ii)  false relative to the context of assessmentwhere B’s standards of taste are operative.

NewRelativist views, which endorse truth-relativism locally for some domain ofdiscourse, stand in opposition to the more traditional view of propositionalcontent (what Cappelen & Hawthorne call “The Simple View”) accordingto which propositions bear truth and falsity as monadic properties (cf.however, MacFarlane 2011a for some resistance to Cappelen & Hawthorne’sclaim that this simple characterization should be regarded as the “received”view.)

:      A key source of philosophical motivationfor relativizing truth in the fashion of New Relativism traces to Lewis’s(1980) and Kaplan’s (1989) foundational work in semantics, according to which sentencetruth is to be understood as relative to a circumstance of evaluation thatincludes world, time and location. New Relativists inherit the formalapparatus of Lewis and Kaplan and add another parameter, but their reasons fordoing so are quite different from the reasons that motivated the framework inthe first place. While Lewis’s and Kaplan’s reasons for “proliferating”parameters were primarily based on considerations to do with intensionaloperators, themore contemporary reasons for adding a judge or standard parameter are often todo with respecting (for instance) disagreement data. (For furtherdiscussion here, see Kölbel (2015)). (Note that “old-style contextualism” canalso be stated in Kaplan’s framework; it involves variation in content withrespect to the context of utterance rather than in truth value with respect tothe circumstance of evaluation).

Kaplan’sview specifically was that the need for particular parameters in thecircumstance of evaluation was a function of the non-specificity of certainpropositional contents with respect to world, time and location (see Kaplan’s(1989) analysis of indexicals). On Kaplan’s view:

Acircumstance will usually include a possible state or history of the world, atime, and perhaps other features as well. The amount of information we requirefrom a circumstance is linked to the degree of specificity of contents and thusto the kinds of operators in the language…. (1989: 502)

JohnMacFarlane, a leading contemporary relativist, writes:

Takingthis line of thought a little farther, the relativist might envision contentsthat are “sense-of-humor neutral” or “standard-of-taste neutral” or“epistemic-state neutral”, and circumstances of evaluation that includeparameters for a sense of humor, a standard of taste or an epistemic state.This move would open up room for the truth value of a proposition to vary withthese “subjective” factors in much the same way that it varies with the world ofevaluation. (MacFarlane 2007: 6–7)

Similarly,Cappelen and Hawthorne write:

Contemporaryanalytic relativists reason as follows: ‘Lewis and Kaplan have shown that weneed torelativize truth to triples of <world, time, location>[’]. … But,having already started down this road, why not exploit these strategiesfurther? In particular, by adding new and exotic parameters into the circumstancesof evaluation, we can allow the contents of thought and talkto be non-specific (in Kaplan’s sense) along dimensions other than world, timeand location. (2009: 10; edited)

Aquestion on which New Relativists are divided, however, is: what contentsare non-specific along dimensions other than world, time and location? It iswith respect to this general question that different families of New Relativismare generated.

The taxonomy weoffer is that a view falls within the category of New Relativism if, and onlyif, the view endorses a truth-relativist semantics (as previously outlined) forutterance tokens in some domain of discourse, such as: discourse aboutpredicates of personal taste [...], epistemic modals [...], future contingents[...], indicative conditionals [...], gradable adjectives [...], deontic modals[...] and knowledge attributions [...]. The motivations for truth-relativism ineach of these domains include various considerations unique to those domains.We consider some of the arguments for New Relativism in four of these domainsin the following sections.

Lui:       This section is not written in a way easyfor the reader’s comprehension except the last paragraph where a definition ofNew Relativism is given. Remember that truth relativism is the two-placepredicate “P is true for X”. Irewrite the positions mentioned in the text within my own understanding (whichmay be wrong):

(1)  Truth absolutism—“P is true” without “for X”regardless of what X is.

(2)  Traditional truth relativism—“P is true for X”where X is a general circumstance of evaluation that includes world, time andlocation. Most likely, X is objective.

(3)  Contextualism—“P is true for X” where X is thespeaker’s standards. X is subjective.

(4)  Semantic invariantism—“P is true for X”provided that X is not the speaker’s standards.

(5)  Newrelativism—“P is true for X” where X is some specific domains of discourse (predicatesof personal taste, epistemic modals, future contingents, indicativeconditionals, gradable adjectives, deontic modals, knowledge attributions,etc.) selected for variousconsiderations unique to those domains. X can be subjective orobjective.  

Clearly,with the exception of truth absolutism, all other positions are the two-placepredicate “P is true for X” where X is delimited differently.

Text:     5.2Truth Relativism and predicates of personal taste

[...]

5.2.1 Lasersohn (2005)and the judge parameter

[...]

5.2.2 Kölbel and Faultless Disagreement

[...]

5.2.3 Moral relativism

[...]

5.3 Truth relativism andepistemic modals

[...]

5.4 Truth relativism andfuture contingents

[...]

5.5 Truth relativism andknowledge ascriptions

[...]

5.6 General Objections toNew Relativism

[...]

5.6.1 Relativism andassertion

[...]

5.6.2 Simplicity

[...]

6. Conclusion

Relativismcomes in a plethora of forms that are themselves grounded in disparatephilosophical motivations. There is no such thing as Relativism simpliciter,and no single argument that would establish or refute every relativistic positionthat has been proposed. Despite this diversity, however, there arecommonalities and family resemblances that justify the use of the label“relativism” for the various views we have discussed. Relativism remains ahotly disputed topic still surviving various attempts to eliminate it fromphilosophical discourse. What is most surprising, however, is the recentpopularity of some versions of the doctrine in at least some circles ofanalytic philosophy.

Lui:       To conclude my commentary, I summarizethe three lessons I have learnt from this SEPentry “Relativism”:

Lesson 1:Typology is a relativizing parameter

Iknow that “the typology of the actor isindispensable to sociological theories” and I said earlier that “I shall for most of the time look upon relativism as typology.” Iam mistaken. Typology is in fact a relativizing parameter because it is adisguise of the often implicit ontology of sociological theories.

Lesson 2:Public stock of knowledge as new relativism

Isaid earlier that “new relativism isnothing but the S-B network of speech, that is, the public stock of knowledge”I am correct, it is just the ““P is truefor X” where X is some specific domains of discourse (predicates of personaltaste, epistemic modals, future contingents, indicative conditionals, gradableadjectives, deontic modals, knowledge attributions, etc.) selected for variousconsiderations unique to those domains.” Every actor picks up from the S-Bnetwork of speech a reference position (“P is true”, a speech) for his own purposes(“for X”; X, some specific domains of discourse).

Lesson 3:Theoretical sociology is the relativizing parameter and sociological theoriesthe relativized variables

Is aid earlier that “recruitment of allknown sociological theories by the theoretical sociologist is linguisticrelativism, with the sociological theories and his theoretical sociology beingrelativized variables and relativizing parameter respectively.” If hesucceeds, then his recruitment realizes the scientific project of sociology asa normal science because “the state ofnormal science is reached for sociology when the semiotic systems of all knownexperimental sociologies can be derived from that of the particular theoreticalsociology so that all of the former are exemplars of the latter as the paradigmof sociology.” 

（Sociological理论大缸第178期）

链接：