SELLARS-HARMAN CORRESPONDENCECORRESPONDENCE BETWEEN WILFRID SELLARS ANDGILBERT HARMAN ON TRUTH{*}ContentsFebruary 26, 1970: Sellars to HarmanMarch 24, 1970: Harman to SellarsNovember 20, 1970: Sellars to HarmanDecember 9, 1970: Harman to SellarsTHE ROCKEFELLER UNIVERSITYNew York, N.Y. l0021February 26, 1970Professor Gilbert HarmanDepartment of PhilosophyPrinceton UniversityPrinceton, New JerseyDear Gil:I have read and reread your insightful and challenging review ofScience and Metaphysics, and have been accumulating notes,I won't say for a reply, but for a first attempt at rising toyour closing challenge. I shall concentrate on the points youraise concerning meaning and truth, leaving to another occasiontopics which, though important, do not touch the very heart ofthe enterprise. Thus I shall not discuss your comments on theavailability and/or adequacy of concepts of thinking-out-loud andits modes as models for the construction of a theoreticalframework which introduce classical thought-episodes. I certainlydid not mean to imply that all the thinking of Jones' Ryleancontemporaries was thinking-out-loud, but only that they did moreof it than we do. Most of their thinking would be, in theterminology of my "Language as Thought and as Communication,"extremely short term proximate propensities to think such andsuch out loud. (I thought I had made this clear in SM.)Also, I have stressed from the beginning that inner thoughtepisodes, though construed on the model of overt speech, are notto be thought of as "words going through ones head." No verbalimagery need occur, and even when it does, the imagery is notitself the thinking, but at most a symptom of it.Moving closer to home, the second paragraph on p. 8 [410] isindeed "truer to (my) intention" than "the view (of inference)deplored above." (I have long been aware that if P implies Q thennot-Q implies not-P.) The distinction between 'rules of action'and 'rules of criticism' plays a central role in myinterpretation of the rule-governed nature of thought. Logicalrules in the primary sense are rules of criticism. Inferring isnot an action. It is not the sort of thing that can be donevoluntarily -- though it may occur in a process which as a wholeis voluntarily undertaken. I shall shortly pick up this generaltheme in the context of "observation".I turn now to the central theme of truth. You write (pp. 8-9[410])"Sometimes he speaks of semantical rules as if they wereprinciples of the theory of truth. At other times he speaks ofthem as if they were principles of the theory of evidence. Thepuzzle is whether he has confused two different sorts ofprinciples, or has rather found a way to exploit connectionsbetween the two sorts of principle."I am not clear why you suppose me to think of semantical rules inmy sense as 'principles' of a 'theory' -- whether of truth or ofevidence. Of course the concept of a semantical rule belongs to atheory, but the theory is a 'theory of knowledge' and deals witha whole family of related topics which include, but are by nomeans exhausted by, truth and evidence. Each type of semanticalrule is in its own way relevant to the explication of each ofthese topics, without thereby becoming a 'principle' of the'theory' of that topic.I do not have much to say about the theory of evidence inSM, primarily because my views on induction andprobability place them in the broader context of a theory ofpractical reasoning, to which the concluding chapter is aprolegomenon. As a result, I was primarily concerned with thebearing of semantical rules on problems pertaining to meaning andtruth. I do, however, touch briefly on the notions of a law-likestatement and inductive reasoning at the beginning of chapter V(sections 4-7), and the discussion of the evolution of conceptualframeworks toward the end of the chapter presupposes (rathercryptically, I am afraid) my general account of theoreticalexplanation.Since I had avoided problems in the theory of evidence, I waspuzzled by the fact that you seem to find them playing apervasive (if confused) role in the argument. When I read "atother times he speaks of (semantical rules) as if they wereprinciples of the theory of evidence," I asked myself 'where?' Ithought you must be referring to my inclusion of L-rules andP-rules of inference among my semantical rules. Imagine, then, mysurprise when on the following page you offer as an example whichshows that "other things that Sellars says makes it seem thatsemantical rules are principles of the theory of evidence afterall," the view, which you attribute to me, that there are"semantical rules connecting observation and thought." "Why," youexpostulate, "bring in observation, if these rules are rules oftruth rather than rules of evidence?" (p. 10 [410]).Now nowhere do I distinguish between 'observation' and 'thought'and discuss their relation to one another. To do so suggests thatan observation is a non-thought which is the awareness of a'state of affairs,' and that the relevant semantical rule saysthat it is correct to think that-p if one is awarethat a certain state-of affairs (perhaps that-p itself, or,perhaps, that it looks to me that-p) obtains. Surely this isexactly the kind of view which (dubbed 'the myth of the given' in"Empiricism and the Philosophy of Mind") I have been vigorouslyattacking since 1948 ("Realism and the New Way of Words"). Toembrace the myth is to treat 'thought entry transitions' asthough they were a special class of inferences, those inwhich the premises are given rather than merely thought tobe the case. My view, on the contrary, is that 'observations' arethemselves thoughts; they are thought-tokens which are correctresponses to the objects which caused them. I deny that there isany such thing as an awareness that-p which isn't a thinkingthat-p. The semantical rules in terms of which the concept ofobservation is to be explicated concern, for example, notthe correctness of thinking 'here is a red object' when one isaware that there is a red object in front of one, or that itlooks to one that there is a red object in front of one, butrather the correctness of the thought 'here is a red object' as aresponse to a red object in front of one. (See also the openingsections of "Language as Thought and as Communication" and pp,10-15 of "Actions and Events.")It is therefore within a framework which I reject that youdevelop the options between which, as you see it, I must choose.Thus you write "everything depends on the exact nature of therules connecting observation and thought. Evidential rules wouldcorrelate stimulation (or how it looks to one) with specificthoughts. Truth rules might correlate the actual (and not justapparent) observation of something with a thought of that thing."(p. 10 [410-11]) Presumably an example conforming to the firstinterpretation would be one in which the datum, itslooking to one that there is a red object in front of one, wouldbe evidential support for the thought that there is a red objectin front of one. Presumably an example conforming to the secondinterpretation would be one in which the observation that thereis a red object in front of one would verify one's thought thatthere is a red object in front of one. Although it is clear thatneither interpretation fits my views, the second, involving as itdoes the idea that what is correct is the thought that an objectis red, on the occasion of stimulation by a red object, is closerthan the first. You sense this, and, therefore, write "presumablySellars has something like the latter sort of rule in mind."[411]Curiously, you give a reference to two passages which, though Iadmit that they are cryptic, are surely incompatible with theframework in terms of which you are confronting me with the abovechoice.This concludes your initial discussion of the question as towhether I confuse truth and evidence. You turn next to my theoryof truth. Here, indeed, the chips are down. The first place at which I began to feel sensitive (like apatient in a dentist's chair) was right at the beginning (p.11[411]) where you ascribe to me the view that "certainrepresentations are true...if and only if they picture particularfacts according to rules of projection in C". Surely I refer towhat is pictured as objects rather than facts.Indeed, already in "Truth and Correspondence" I had emphasizedthe distinction between the pseudo-relation of 'correspondence'between true statements and facts (a pseudo-relation because itturns out that true statements are facts) and the genuinerelation of picturing which holds between certain statements asitems in the spatio-temporal-causal order and the objects withwhich they are correlated by the rules of projection of thelanguage. The schema for picturing is not(statement) pictures (fact)but rather(object) pictures (objects).But, then, since the word 'fact' is used in different ways bydifferent philosophers, there was, as yet, no serious cause foralarm.The remarks on p. 12 [412] are misleading in that they do notreflect my views on the nature of law-like statements. (They arematerial rules of inference.) On my view, also, the concept oftwo worlds which agree in their histories, but differ in theirlaws, is incoherent. But this is because attributes and laws arenot conceptually independent, as traditional empiricism holdsthem to be. Thus geometrical attributes are not conceptuallyindependent of geometrical principles, nor micro-physicalattributes of micro-physical principles. On my view, law-likestatements are material rules of inference of the form (roughly):from 'K is a finite unexamined class of As,' it iscorrect to infer 'n/m K is B.'An open quantified counterpart of the result of afinite application of this rule would be a statement ofthe formn/m (of all) A is B.And, where n/m is 1, (x) Ax --> Bx.Law-like statements themselves would, as meta-linguistic rules,be "determined" by the evidence, in the sense that the adoptionof the rule is vindicated by practical reasoning in which thedescription of the evidence occurs as a premise. Needless to saythe equivalence'it is correct to infer . . . from ---' is trueiff it is correct to infer . . . from ---.holds as in all cases of truth. But the semantical ruleswith respect to which the truth (semantic assertibility) of'it is correct to infer . . . from ---' is to be understood,are in no simple way related to the truth of atomicrepresentations. For the rules in question are those whichdefine inductive reasoning, and fall outside the recursivespecification of truth conditions for molecular and quantifiedmatter-of-factual statements.Thus when you ascribe to me the view that atomic representations are true if and only if theycorrectly picture the facts (sic), whereas the truth of otherrepresentations is completely determined by the set of trueatomic representations (p. 13 [412])I not only have serious reservations about the first clause,since the term 'fact' tends to short-circuit theories oftruth, but would categorically reject the second clause,if it is intended to apply [to] all truths, (ethicalprinciples, law-like statements, theoretical principles) otherthan true atomic representations. On the other hand, it would bedifficult not to agree that the truth of truth-functions(including quantificational derivatives) of atomic sentencesis so determined. Yet even here the relation of inferentialconnection to the truth of basic sentences must not be overlookedin assessing the force of "is determined by".I come now to the heart of the matter, your critiqueof my critique of "Carnap-Tarski semantics." Much of whatyou have to say seems to me to rest on a misunderstandingfor which I must share at least part of the responsibility.My attack is directed not against the idea of a recursivespecification of truth conditions (how else would one explain therelation of the truth of 'p or q' to the truth or falsity'p' and 'q'!), but to the idea that a recursive specificationof truth conditions provides an explication of theconcept of truth. I was looking for an account of theintension of the predicate 'true' and, hence, wasconcerned to distinguish carefully between the intensionof 'true' and the conditions which sentences of variouslogical form must satisfy for this predicate to be (truly)applicable to them. One is tempted to put this by invoking thetraditional contrast between the 'meaning' and the 'criteria' oftruth, but 'criteria' is an accordion word, and its use mightcontaminate the discussion with the problem of evidence discussedabove.The same motivation guided my critique of Carnap'saccount of such syntactical predicates as 'sentence (ofL)' in my Carnap volume essay. The point at issue is,in a familiar sense, philosophical rather than substantive.Thus there is an important sense in which what I reject isnot Carnap-Tarski semantics, but rather the Carnap-Tarskiphilosophical interpretation of Carnap-Tarskisemantics.It is probably my fault that you did not pick up thesignificance of my claim to give an account of the intensionof 'true.' Otherwise it might have occurred to you that Icould agree with the Carnap-Tarski recursive account oftruth-conditions (and satisfaction-conditions), withoutagreeing that they had captured the intension of either'true' or 'satisfies.' You would have then been prepared to findme claiming that an adequate explication of 'true' and'satisfies' must break out of the family of semantical conceptsand relate them to such 'pragmatic' concepts as'would be a correct perceptual response to,' i.e. to notionspertaining to the observation-inference-action game at allits levels.The above general remarks should provide the background againstwhich the following more detailed comments might fall into place.Notice, to begin with, that whereas I write in terms of acontrast between the intensions of 'true' and 'true of'(or its converse 'satisfies') and the conditions whichgovern the applicability of these predicates, you blur thisdistinction by formulating the issue as one which concerns the"characterization" of truth. Thus you write (p. 14 [413]) . . . because it attempts to offer a recursivecharacterization of truth, Tarski's theory can also beinterpreted as providing a correspondence theory somewhat similarto Sellars' picture theory of truth.The preceding remarks should make it clear that from my point ofview, Tarski does not "provide" a "correspondence theory somewhatsimilar to Sellars' picture theory of truth," though it iscompatible with a picture theory of the truth ofmatter-of-factual statements. Such a genuine picture theory wouldgive a pragmatic account of the conditions which a basicmatter-of-factual statement must satisfy to be "semanticallyassertible," and a recursive account of the semanticalassertibility of molecular compounds of such statements. I cansay 'pragmatic', because the analysis of the conditions which abasic sentence (I should now begin to speak in terms ofstatements) must satisfy in order to be 'semantically assertible'will be in terms of such notions as 'correct perceptual responseto object Oi (or objects Oi, Oj)' and 'is an appropriate tensedcounterpart of statement S.'In the passage which follows the sentence quoted above,you write:Things are complicated by the fact that Tarskienvisions a recursive characterization (sic) of satisfaction aswell as of truth, where Sellars envisions a non-recursivecharacterization (sic) of truth apart fromsatisfaction.Without running the point into the ground, let me stress againthat my aim was to give a non-recursive definition oftruth, i.e. analysis or explication of 'true'. I am quite happywith the idea that truth-conditions are to be recursivelyspecified. This does not mean, of course, that alltruth-conditions (e.g. truth conditions for ethical statements)belong in the same recursive hierarchy, as would be thecase if all sentences were truth functions of atomicmatter-of-factual sentences, or quantifications of openmatter-of-factual sentences. Thus, I do not envision anon-recursive characterization (as contrasted withexplication) of truth. What, then, of satisfaction? Do Ithink that to characterize truth we must make use of the'relation of satisfaction?' The answer is in the affirmative. Thetruth conditions of quantified statements are, indeed, to bespecified recursively in terms of satisfaction. This, however, iscompatible with the fact, on which I have laid great stress, thatthe focal concept of semantics is that of truth, the cash valueof which is to be found in the 'truth move' illustrated by 'Snow is white' is trueso, snow is whiteReflection on the following two examples should help me makethe point I wish to make about satisfaction:'It is not day or the sun is shining' is trueso, it is not day or the sun is shining'(x) x is a man --> is mortal' is trueso, (x) x is a man --> x is mortalThe identity of sense of 'true' as it occurs in these twoexamples is exhibited by the common form of the sequence. Thisidentity of sense, however, goes along with the fact the truthcondition of 'either it is not day or the sun is shining' isrecursively specified in terms of the formulas'not-p or q' is true iff 'not-p' is true or'q' is true'not-p' is true iff 'p' is false.Similar considerations hold in the case of the second example.But before I elaborate this point I shall examine thefollowing passage from your review,. . . this brings us to Sellars' second reason for denying that"Tarski-Carnap semantics" provides any sort of correspondencetheory of truth. Sellars claims (in effect) that the relation ofsatisfaction, which plays an essential role in Tarski's accountof truth for quantified sentences, is not really a semanticrelation at all. That is, he denies that satisfaction is arelation between language and the world. [413]Now I certainly do not deny that satisfaction is a semanticalconcept. And I certainly do not deny that when we get to thelevel of basic sentences our account of truth conditions mustinclude a relation between language and the world. What I deny isthat satisfaction, as characterized in Tarski-Carnapsemantics, is that relation. It is close by and near to thatrelation, but not identical with it. Leaving aside for a momentthe considerations which focus attention on a concept ofsatisfaction which applies to sequences of objects, an object (oran n-tuple of objects) satisfies an open basic sentence if andonly if (an appropriately tensed) closure of that sentence whichcontains designations of the object or objects would betrue. Again, a present tensed closure would be true if andonly if it would be a correct perceptual response to (i.e. wouldpicture) the object(s) (Compare my concept of a 'verifying token'in "Realism and the New Way of Words.") Other differently, butappropriately, tensed counterparts would picture the object (orobjects) by virtue of their relation to such a verifyingtoken.The above considerations point to a more precise formulation ofthe point at issue between the theory of truth I am defending,and the standard philosophical interpretation of the Tarskiformalism. Granted that the truth of matter-of-factual statementsinvolves at some point a direct relation between languageand the world, and that statements which are not true by virtueof a direct relation to the world are true by virtue of beingindirectly related to it in a way which can be specified byrecursion -- i.e. granted all this, is this language-worldrelationship captured by the intention of either 'true' or'true of' or 'satisfies?' My answer is no. The answer ofthese who hold the standard interpretation of the Carnap-Tarskiaccount -- most recently Donald Davidson -- is yes. Thus,according to the latter "...the property of being true has beenexplained... in terms of a relation between language andsomething else . . . the relation, satisfaction . . ."{1}Very well, then, what is satisfaction if it is nota relation between language and the world? The clue to theanswer lies in the fact that the concept of the satisfactionof basic open sentences provides the basis for a recursiveaccount of the truth of quantified sentences just as,abstracting from considerations of quantification, the concept ofthe truth of basic closed sentences provides the basis for arecursive account of the truth of their molecular compounds.Tarski, indeed, goes on to show us how it is possible to unitethese recursive structures by defining the truth of a basicclosed sentence in terms of satisfaction. Thus, if satisfactionwere, indeed, a relation between open sentences and objects (moreaccurately, sequences of objects), we would, indeed, have beenshown that the concept of truth is at bottom the concept of arelation between language and the world.Now we can imagine a philosopher who is a 'finitist' -- not in the sense that he abjures the Cantorian paradise, butin the sense that he interprets references to varieties ofinfinite collections to be non-perspicuous references tovarieties of open ended mathematical rules -- arguing thatTarski's recursive formalism can be restructured to take thetruth of basic closed sentences as primary, and to treat thetruth of quantified sentences as special case of the truth ofmolecular compounds. He could be expected to argue that onlyif the truth of basic closed sentences could be shown to be (notjust 'involve' -- as of course they do) a relation betweenlanguage and the world, could something like the traditionalcorrespondence theory of the meaning of 'true' have beenestablished. Of course in such a hypothetical (and heretical)reconstruction, the semantical predicates 'true of' and'satisfies' would be definable in such a way that, for example,'fx' is true of a iff 'fa' is truewould be true by definition. But this would provide grounds forsaying that the truth of 'fa' is a relation between 'fa' andsomething non-linguistic only if we already knew informally, as acriterion for the adequacy of the definition, that 'true of'stands for a relation between open sentences and objects. If, topush the matter one step further, it is pointed out that theabove example should read 'fx' is true of a iff 'fa' is true and 'a' denotes aour philosopher would claim that we are committed to the ideathat 'true of' stands for a relation between open sentences andobjects if and only if either the truth of 'fa' is (notjust 'involves') a relation between 'fa' and somethingnon-linguistic, or 'denotes' stands for (not just'involves') a relation between a linguistic and non-linguisticitem.I have introduced this hypothetical position to focus attentionon the manner in which a number of theses which I defend inSM stand or fall together. Thus:(1) I insisted that the truth of a basic sentence (like alltruth) is semantic assertibility. The truth of basicsentences, however, involves, indeed is grounded in, theexistence of a relation of picturing between tokens of thesentences and the objects to which they refer.(2) I insisted that 'denotes' does not stand for a relationbetween a linguistic expression and an object. On the other handthe denotation statement would not be true unless the expressiondid stand in complex matter-of factual relations to theobject.(3) I insisted that the 'stands for' in 'f' stands for __________ does not stand for a relation between a linguistic andnon-linguistic item, although the indented statement would not betrue unless 'f' stood in complex matter-of-factual relations toverbal behavior and non-linguistic objects.(4) I insisted that the contexts'fx' is true of aa satisfies 'fx''b' denotes aare, in Frege's sense oblique. In scholastic terms, theexpression 'a' in the above examples is used in secondintention.All of this is by way of comment on an intriguing passage inDavidson's paper in which, commenting on 'correspondence to fact'theories of truth, he writes:One well-explored consequence is that it becomesdifficult to describe the fact that verifies the sentence exceptby using that sentence itself. The other consequence is that therelation of correspondence (or "picturing") seems to have directapplication to only the simplest sentences (Dolores lovesDagmar). This prompts fact-theorists to try to explain the truthof all sentences in terms of the truth of the simplest and hence,in particular, to interpret quantification as mere shorthand forconjunctions or alternations (perhaps infinite in length) of thesimplest sentences. The irony is that, in so far as we can seequantification in this light, there is no real need for anythinglike correspondence.That I am not a 'correspondence to fact' theorist should,by now,be clear. Yet if 'fact' is replaced by 'objects' I am a correspondence-to-objects theorist. For it is an essentialfeature of my theory that although 'true' does not meancorresponds to objects, but rather semantically assertible,nevertheless, correspondence to the relevant objects is anecessary and sufficient condition of the semanticalassertibility of a basic statement, and this would be so even ifquantificational theory could, in principle, be treated as anextension of propositional logic, i.e. even if our recursiveaccount of truth conditions could have as its basis theenumeration,'fa' is true iff Tom is tall...'gb' is true iff Dick is thin...Where, given PMese syntax, and the dictionary,'a' des Tom'b' des Dick.. . 'f' des tall things'g' des thin things...we could also make such finer-grained statements as that'fx' is true of Tom iff Tom C tall things'gx' is true of Dick iff Dick C thin things...and recursively specify the truth conditions of molecularsenses (including quantified sentences) belonging to thelanguage thus characterized.Against this background, let me return to your comments.After pointing out thatStrictly speaking, one should say that an opensentence is satisfied by a sequence of objects [414](by which you are, I take it, calling attention to the factthat according to Tarski the immediate concept we need toprovide a recursive account of the truth of quantified sentencesrelates open sentences to sequences of objects, and is to bedistinguished from a more basic concept of satisfactionaccording to which an object or n-tuple of objectssatisfies, i.e. belongs to the extension of, an open sentence,where this latter concept is to be taken a semanticalprimitive.) You continue,for our purposes we may think of a predicate expression assatisfied by certain objects when it is true of them.[414]Here you are, I take it, calling attention to the fact that thesemantical primitive of Tarski's account is the idea of an objector n-tuple of objects belonging to [the] extension of an opensentence. In other words, the semantical primitive of Tarski'saccount is the concept of an open sentence being true ofan n-tuple of objects. If one begins to feel that the Tarskiaccount is moving in a circle -- usefully, however, and ina circle which is large enough for its purpose -- onemight also begin to feel that the connection between Tarski'ssemantical primitive and the world, which connections involveslearned patterns of verbal behavior, remains to be explicated.This is what my account attempts to do. That I find thefocal point of the connection to be at the level of 'true' ratherthan 'true of' does not mean that I minimize the role of'true of' in explicating the semantics of quantification. Istress 'true' because here is where the matter-of-factualrelationship of picturing, definable in the framework of languageas a rule-governed phenomenon in rerum natura, joins upwith the 'truth move' which is characteristic of all modes oftruth and which is exhibited by the schemaThat-p is trueso pI continue with your text.. . . for our purposes we may think of a predicate expressionas satisfied by certain objects when it is true of them.Thus, "is wise" is satisfied by Tarski. Now Sellars says, ineffect, that what I have just said in the preceding sentence onlyappears to assert a relation between language and the world. Heargues that my claim really asserts a relation between theexpression "is wise" and "Tarski" and he suggests that it wouldbe more accurate to write it in any of the following (for himequivalent) ways:"is wise" is satisfied by "Tarski""is wise" is true of "Tarski""Tarski"^"is wise" is true"Tarski is wise" is trueHe concludes that "Tarski-Carnap" semantics does not show howlanguage is related to the world. (p. 15 [414])Now I do indeed claim that "Tarski-Carnap" semantics does notshow how language is related to the world. I criticize itsclaim to have explicated the tie between language and theworld. My present purpose, however, is to reply to your criticismof my treatment of satisfaction.The first point is a relatively trivial one. I do not hold that"Tarski is wise" is trueis a 'way of writing'"is wise" is satisfied by (or true of) "Tarski"I hold, simply, that satisfaction and truth of aredefinable in terms of 'true'. You criticize this claim as followsNotice that Sellars' analysis works only if every object thatsatisfies any expression has a name. I do not think that this is true, but to show it I must mobilizefeatures of my account of meaning and truth on which you do nottouch. It is an essential feature of my philosophy of languagethat the nature of quantification and the role of variables is tobe explicated in terms of the concept of an ideal language whichcontains names of all objects, and predicates for all basicattributes and relations. Such a language I called, in "Realismand the New Way of Words," the language of omniscience. In suchan ideal language there would be the resources for constructingbasic sentences which adequately picture all objects. I arguedthat we should view our language as a "schema" of such an ideallanguage. The cash value of this, for my present purposes, isthat we should view the variables of our language as taking fortheir substituends not only the constants we now have, butthe constants which are to be introduced in accordance withrational methods. (See SM, V, 66). Looked atbehavioristically the matter-of-factual connection betweenlanguage and the world includes acquirable as well asacquired patterns of verbal behavior. It is the verynature of language to have a reach which exceeds its grasp. Thisfeature belongs not only to object language variables but tometa-linguistic variables as well. It is a feature of naturallanguages which recurs at all levels.This feature of quantification enters as follows into my accountof 'true of' and 'satisfies.' Abstracting from my concern withinter-linguistic identity of roles in the conceptual game ofobservation-inference-action, and limiting myself to roles asembodied in English sign-designs, and emphasizing that English isnot be taken as English 1970 but as the continuing Englishof the English speaking scientific cammunity, the following(regimented) account expresses my views. Let 'IND,' 'PRED,''SENT,' etc. be meta-linguistic variables which take quotedexpressions as substituends. With these resourses we can say notonly'is wise' is true of 'Tarski' iff 'Tarski is wise' is truebut(E IND) 'is wise' is true of IND iff (E IND)IND^'is wise' is true.I need no more be able to replace(E IND) IND 'is wise' is trueby a disjunction which begins'Harman is wise' is true or 'Tarski is wise' is true . . .then I need be able to replace(E x) x is tallby a disjunction which beginsTom is tall or Dick is tall . . .Nor does'x' ranges over a, b, c . . .formulate a 'ranging relation' between the variable 'x' andnon-linguistic entities. It is rather the material mode for'x' has as substituends 'a,' 'b,' 'c,'...Note that it would be a serious mistake to suppose that the twoindented statements are equivalent to'x' ranges over 'a,' 'b,'. . .The formal and material modes of speech must not be mixedtogether. I have put this elsewhere by saying that contexts suchas 'ranges over . . .' already play a quoting role, so that tofollow them with quoted expressions moves them up one higherlevel than is intended. Actually the same is true of thecontext 'true of . . .' That is why it is best to underline theexpression following it rather than put the expression in quotes.Yet putting the expression following 'true of' in quotationmarks, though technically incorrect, does serve the purpose ofemphasizing that the expression which properly follows 'true of'is being mentioned (autonymously) rather than simply used. If the interpretations of range statements as materialmode of speech for substitution statements seems shocking, it isbecause we tend to overlook the fact that the right handside involves a rule of construction. In the above model the ruleis based on the alphabet. In a natural language it would involvethe rules concerning the introduction of referring expressionsinto the language. These involve something like a coordinatesystem, and can initially be characterized as syntactical andpragmatical rules governing the use of such a system. Thus, tooversimplify we can envisage the use of numerals as subscriptsmaking referring expressions out of variables.'x' ranges over x1, x2, x3 . . .'x' has as substituends 'x1,' 'x2,' 'x3,'. . .Once again, as in the case of other semantical concepts, thesemantical concept of 'ranging over' is connected with apragmatic notion pertaining to patterns of verbal behavior inrelation to extra-linguistic objects, but is not thisnotion.Strictly speaking, of course, objects are not locations, but havelocations. This is why the substituends for 'x' in 'x is wise'are not as immediately available as are the terms in anarithmetical sequence of which the rule is known. As I pointedout above, the potential substituends for the variables of acontinuing language are potential with respect to the scientificenterprise as a whole, and not with respect to anything as simpleas the rules governing a coordinate system.You write. . . Tarski introduced to the notion of satisfaction in order to give truth conditions for quantified sentences. He wouldnot have had to do so if he had been able to assume that allobjects in the range of a quantifier have names in the language,for if they do one can make do with so-called substitutionalquantification. [And indeed, Sellars' version of quantification]is the substitutional variety.(15-6)[414].I have always been uneasy about the phrase 'substitutionalvariety of quantification,' which implies that there are two'varieties' of quantification, 'substitutional' and 'objectual'(or, as you put it, 'referential'). As I see it there is justquantification, and the distinction between 'substitutional' and'objectual' is a not too happy way of contrasting twotheories of the sense of quantified statements. My versionof quantification is 'substitutional' only in the sense that(1) I interpret statements about the range of variables along theabove lines, in keeping with my general approach to semanticalstatements.(2) I regard the following statements as quantificational inexactly the same sense:(Ec) c(p,q)thus, or(p,q)(Ef) Tom is fthus, Tom is tall(Ex) x is tallthus, Tom is tallThis parallels the above account of the range of 'x'; thus 'c' ranges over connectiveshas the sense of'c' has as substituends, 'and' 'or' 'not' . . .and'f' ranges over propertieshas the sense of'f' has as substituends 'red' . . . 'square' . . .I, therefore, see no reason why I cannot avail myself of Tarski'smethod of giving truth conditions for quantified sentences. Iquite agree with him that it is necessary to use the concept ofsatisfaction to specify these conditions. I see nothingincompatible with this in my analysis of satisfaction. You writeSellars is just wrong when he says that the relation ofsatisfaction in "Tarski-Carnap semantics" does not relatelanguage and the world. On the contrary it only works because itdoes relate language and the world.[414]On the contrary, it works because of its formal properties andnot because of the Tarski-Carnap conviction that 'true of'stands for a relation between linguistic andnon-linguistic entities. You admit, in the paragraph whichfollows the above, that I can "give truth conditions . . . forsubstitutionally quantified sentences," though I do not make useof "the notion of satisfaction," although you insist that Tarskican "give truth conditions for referentially quantifiedsentences" because he makes use of "the notion of satisfaction."I reply that Tarski and I are both making use of the samesemantical concept of satisfaction, and giving truth conditionsfor the same quantified sentences -- but are givingdifferent philosophical accounts of what we are doing. Thedifference between us is not a semantical one, but ameta-semantical one. It is in this spirit that I accept yourolive branchSo perhaps no real objection to Sellars' theory arises here.[415]Now I come to a comment which really perplexes me:Tarski requires that a truth characterization satisfy conditionT{2}referred to above. Sellars does not require this.[415]You give no grounds for this claim. It apparently has somethingto do with my treatment of law-like statements, but since I havelittle to say about this in SM, and since you give no hintas to what feature of my treatment motivates your remark, Isimply do not see what you are driving at. What I do say aboutlaw-like statements (V,5) characterizes themas material principles of inference. But why equivalences of theform'That something is f materially implies that it is g' is true iffthat something is f materially implies that it is g.should be exceptions to condition T, I completely fail to see.You do, however, go on to make another point about the place ofcondition T in my account which does touch closer to home. Younote that whereas I have quite a bit to say about the sense inwhich 'it is true that Carnap is wise' implies 'Carnap is wise,'I offer no ". . . argument to show that Carnap is wise onlyif it is true that Carnap is wise." Thus, you conclude, I "donot offer any account of why Carnap is wise if and only ifit is true that Carnap is wise.This is a fair criticism of the argument of the chapter on truth,even though at no point in that chapter did I "pretend toexplain" why Carnap is wise if and only if it is true thatCarnap is wise. I should have had something to say aboutthe sense in which 'Carnap is wise' implies 'it is true thatCarnap is wise.' My failure to do so was simply anotherexpression of the fact that my disagreement with what I(unfortunately) called "Carnap-Tarski semantics" wasphilosophical rather than formal.Let me attempt to remedy this omission, or, more accurate tosketch the general line such a remedy would take. The point is animportant one, because it requires an explicit discussion of theconnection between the concept of truth and the concept ofsemantical rule. Thus, if I had filled this gap in my discussionof truth, the question of whether my semantical rules are rulesof truth [or rules] of evidence, to which the first part of thisletter was devoted, would never have arisen.[editor's note: The following passages were amended by Sellars inhis Nov. 20 letter to Harman.]The problem is most pressing with respect to basic sentences.Yet, to fix our ideas (as Nagel says), it is worth looking at aless problematic case. In what sense does 'p or q' imply 'it istrue that p or q?' Or, to translate this in terms of my analysis,but omitting dot-niceties, in what sense does 'p or q' imply " 'por q' is semantically assertible (with respect to relevantrules)?" Or, again, what gives kosherity to the sequence 1. p or q2. 'p or q' is SA (semantically assertible).Now, to inscribe the design p or q in a language-usingframe of mind, implies that it can be classified as a token ofsome expression or other. And, indeed, the question presupposesthat inscription 1 can be classified as a 'p or q.' Furthermore,although we want to say that in an extended sense the implicationwe are seeking to understand holds in the case of all tokens of'p or q' whether they occur in play rehearsals, examples,tape-recorder-play-backs, tall stories, etc., we distinguish asprimary the candid, cognitively motivated, use oflanguage. Assuming this primary context, we consider thealternatives1 conforms to the relevant semantical correctness (isSA)1 does not conform to the relevant semantical correctness (is notSA)So far we have 1: p or q2: 1 is a primary token of 'p or q'3: 1 is SA iff 'p or q' is SA.Now the statementp or q, but 'p or q' is not SAis, in the sense which Moore attempted to analyze incoherent.Thus if we are prepared to stick with 1, the derivation continues4: 1 is SA (pragmatic implication)5: 'p or q' is SA.So far our sequence exhibits the connection between thesemantical assertibility (truth) of 'p or q' and the semanticassertibility (truth) of " 'p or q' is SA." To show how theformulation of this connection is justified, we continuethe derivation as follows:6: 5 is a primary token of " 'p or q' is SA"7: 5 is SA iff " 'p or q' is SA" is SA8: 5 is SA (pragmatic implication)9: " 'p or q' is SA" is SAThis derivation, as before, contains a pragmatic step.Nevertheless it is conceptually binding and explains theincoherence of the sequencep or q'p or q' is not SA (true)and hence the sense in which 'p or q' implies 'it is true that por q.' And, after all, if I am right that the 'if' direction ofthe equivalenceit is true that p or q iff p or qinvolves a pragmatic consideration (the 'truth-move') it shouldnot be surprising that pragmatic considerations enter into the'only if' direction as well.But what, you may ask, are the 'semantical correctnesses' towhich you refer? 'Surely the above account is empty until theyare specified. To this the answer is that in part they arecorrectnesses which relate to the function of the connective 'or'and are reasonably unproblematic. But in part they concern thepropositional constituents of 'p or q' and raise the puzzles of'truth' all over again. To resolve them we must first take intoaccount the recursive characterization of truth conditions whichis so important a feature of Tarski's account. This tells us thatit is true that p or q iff it is true that p or it is true thatqThus, given that 'p', for example, is a basic sentence, thistakes us to the problem of the equivalenceit is true that p iff pAssuming, as before that 'truth-move' strategy takes care of 'if'direction, the problem remains of explicating the sense in which'p' implies 'it is true that p.'[editor's note: This is the end of the amended paragraphs.]The strategy is the same, but this time we come face to face withthe problem: what are the relations between basic sentences andobjects (not facts) which define the truth conditions ofthe sentences, and, hence, indirectly, the truth conditions ofmolecular and quantified sentences at this level?In this context, of course, tenses (and demonstratives generally)become important, and attention must be paid to the fact thattokens of differently tensed sentences can "make the samestatement" i.e. translate into one and the same tensed sentencewhich is appropriate to our use now.I sketched a classification of the semantical rules relevant toour problem in the table in the end of chapter V. Of particularimportance are those belonging to categories II, (a), (alpha) andII, (a), (beta). The fundamental principle is that a basicsentence, in the appropriate tense for our use now, is true ifand only if a token of the corresponding present tensesentence would be (would have been) a correct language entry(perceptual) response to a certain object (or objects). Let uscall such a correct response a 'verifyingtoken' of the basic sentence. Its correctness involves (a) thatthe singular term or terms it contains be appropriate; (b) thatthe predicate be appropriate.That a perceptual response token of 'this is red' be a verifyingtoken for 'a was red' requires that semantical rules connect thesingular term 'a' with an object located at the place and timewith respect to which the perceptual response was made. This'connection' of the singular term 'a' with an object is not thepseudo-relation of denotation, but is connected with the lattermuch as the truth condition of a specific proposition isconnected with its character of being true. It is an empiricalconnection in the causal order between 'a' and other expressionsin a schematic and developing world story, a sub-set of thesentences which are here-now perceptual responses. Theidentification criteria for 'a' must connect it with othersentences which, in present tense form, would have been correctperceptual responses to a and to other objectsspatio-temporally related to it. The application of the languageessentially involves the application of a (schematic) world storyformulated in that language (compare V, 30). A furtherexploration of this point would tie it in with the discussion ofthe range of variables for individuals in an earlier paragraph ofthis letter.Again, that a perceptual response token of 'this isred' be a verifying token of 'a is red' requires that tokenswhich contain the predicative expression 'red' be correctresponses to red objects. Here is where my interpretationof the subject-predicate distinction, and the status ofabstract entities, becomes important. For if they arenot taken into account the confusions of fact-ontologiesbecome, as I see it, unavoidable. My analysis is intendedto explain, without reference to accessibility to quantification,the sense in which predicates are syncategorimatic expressions,and, together with my account of abstract singular terms, toexplain why quantification over predicate variables does notcommit one to a platonistic ontology.A token of 'fa', viewed in terms of what might becalled its depth-grammar, is not an 'f' concatenated withan 'a' but rather an f*/'a'/, where an f*/'a'/ is an 'a'which is concatenated with an 'f,' just as a white/dog/ isa dog which is white. The second requirement, then, is tothe effect that red* singular terms are appropriate responsesto red objects.The importance of this account is the fact that itkeeps us from looking for an object 'red', and by gettingus away from the name-object model, enables us to see how thesemantical rules for the predicate 'red', i.e. the rules forbeing a red* singular term can essentially involve itsrelations to other color predicates and to predicates of thespace-time family. It enables us to see how these rules can be,as I put it in "Realism and the New Way of Words," conformationrules. It is but a step from here to argue the involvement oflaw-like statements in the meaning of predicates.I know that I have barely scratched the surface of the concept ofa verifying token. But the above might be useful by suggestinguseful questions. Let me continue, then, by saying that verifyingtokens are primary pictures of objects. Other tokens are picturesif they are appropriate transforms of primary tokens.Notice that the correctness of a verifying token (primarypicture) is, in the terminology of "Language as Thought and asCommunication," a matter of its conforming to ought-to-bes,rather than ought-to-dos. For perceptual responses are, as such,involuntary; passions rather than actions. It is ultimately interms of such ought-to-be correctness, that the truth conditions,i.e. conditions of the semantic assertibility, of basic sentencesis to be defined. Roughly'fa' is true (SA) iff 'fa's picture aThe right-hand side of this equivalence formulates the truth condition for 'fa.' This time, however, unlike the other cases inthe recursive chain, the condition directly involves the relationof picturing which, unlike concepts belonging to the familyexplored by classical semantics, (stands for, denotes, extension,intension, reference, truth, truth of, etc,) is definable innaturalistic terms, thus word-word and word-world uniformities.To be sure, the uniformities are controlled, along the linesdiscussed in "Language as Thought and as Communication," byought-to-bes (principles of criticism), which are, themselves,related to ought-to-dos (principles of action) relating to verbalbehavior. But picturing itself is a natural, not a normative,relationship. I have explored the 'causal' force of normativelanguage in a number of papers, but it would be folly to touch onit here.This concludes what I have to say (on this occasion) in reply toyour critique or my critique of "Tarski-Carnap semantics". Ishall close with brief comments on the remaining paragraphs ofyour review.My argument to the effect that the world as conceived in theManifest Image is, in the Kantian sense, phenomenal, hinges noton considerations pertaining to sets, sums, and slices, as youseem to imply [417], but on considerations pertaining to the wayin which sense qualities are involved in the constitution ofthese objects. It has nothing to do with the question whether onchoice of primitives for a given framework is better thananother. Incidentally, your use of [the] term 'colorless' on p,19 [417] trades on an ambiguity. It may mean 'looks like clearwater', and stand in this sense for a visual quality, or it maymean 'not having a visual quality at all.' Atoms are notcolorless in the first sense, but rather in the second.I have nowhere said that the Scientific Image doesn't includeclusters, configurations, sets, wholes, slices, etc. of atoms northat it doesn't include atomic events, states, processes,interactions, etc., etc. My general views on ontologicalcategories applies to the Scientific as well as to the ManifestImage. I have said it doesn't include chairs, tables,etc., as conceived in the Manifest Image (see above). Itdoes, however, include 'successor concepts' of the relevantManifest concepts. Thus, as I point out at the end of Chapter V,there is a sense in which the chairs, tables, etc, of theManifest Image really exist.As for your concluding reprise of the rules of truth, rules ofevidence theme, I think that my letter has directly or implicitlycovered this ground. If not, however, I hope that my remarks havenot been so baffling that they have failed to advance our (Ihope) continuing dialogue.As ever,Wilfrid SellarsWS/mf[Contents]Princeton UniversityDepartment of Philosophy1879 HallPrinceton, New Jersey 08540March 24, 1970Professor Wilfrid SellarsThe Rockefeller UniversityNew York, New York 10021Dear Wilfrid:Thank you for the comments on my review of Science andMetaphysics. I think I finally see the difference betweenwhat you are trying to do and what Tarski and Carnap were tryingto do. You want to say what "true" means, whereas they attemptedto give necessary end sufficient conditions for truth in one oranother particular formal language.I think the difference can be expressed more clearly if what youare doing is compared with what Quine does in section 4 of "TwoDogmas of Empiricism." Quine argues that Carnap does not explainwhat "analytic" means. He points out that Carnap defines"analytic-in-L0" for a particular language L0 but never defines"analytic-in-L" for varying L. Thus Carnap nevercharacterizes analyticity in general. Similarly, you might say,the semantic theory of truth never defines "true-in-L" forvarying L; at best it defines "true-in-L0" for a particularlanguage L0.Quine points out that any adequate account or analyticity ingeneral would have to refer to "mental or behavioral or culturalfactors." Similarly, you point out that the same holds for anyadequate account or truth in general. In particular, thesemantical rules of truth must have something to do with the useof language, If I understand you, you go on to suggest that theserules are rules in accordance with which users of the languagebehave.This way of looking at what you are doing clears up some of myconfusions about what you say about Tarski. For example, on page21 you say that Tarski's account moves in a large circle. I wasoriginally puzzled by this, since Tarski's account seems no morecircular than any other recursive definition. Your point is, notthat Tarski's account is circular as an account of truth-in-L0,but that it becomes circular if turned into an account oftruth-in-L for variable L.Similarly, I take it that when you deny that Tarski's relation ofsatisfaction relates objects with linguistic expressions,what you mean is that it cannot do this if it is treated as arelation introduced into an account of truth in general. For itis obvious that as a relation introduced simply to account fortruth-in-L0 satisfaction does relate (sequences of)objects to linguistic expressions. It is simply defined to dothat. Suppose a relation R is defined as follows:"Rxy =df x is wise and y is the expression 'is wise'."R is defined to be a relation between objects and alinguistic expression. Tarski's relationsatisfaction(-in-L0) is simply a more complex version ofR.Quine specifies what he takes to be the relevant "mental orbehavioral or cultural factors" in Word and Object. Theyturn out to be a meager lot, definable in terms of a speaker'sdisposition to assent to or dissent from sentences as the resultof varying sensory stimulation.What you count as relevant "mental or behavioral or culturalfactors" differ in two important respects from what Quine countsas such factors. For one thing, you think there are more factors:in addition to perceptual responses, you mention inferenceand the effect of practical reasoning on action.Furthermore, where for Quine a perceptual response would be aresponse to certain stimulation, for you it is a response to aperceived object that causes the response. And this second pointis a special case of a more general difference between Quine'sapproach and yours. An obscure way to put this (rightly condemnedin your letter) is that Quine thinks that speakers of a languageact in accordance with rules of evidence, which give meaning totheir words, whereas you think that speakers act in accordancewith rules of truth, which give meaning to their words.Thus for Quine, what speakers learn when they learn "Rabbit" isto assent to "Rabbit" given certain conditions of stimulation.For you, speakers learn to assent to "Rabbit" given theperception of an actual rabbit. Or, and better, for you, speakerslearn to assent to "That is a rabbit" given the perception of anactual rabbit. Such a response pictures a perceived rabbit as arabbit.(I went wrong in my review when I spoke of responses toobservation where I should have spoken of perceptual responses. Iwent wrong again when I spoke of picturing the facts where Ishould have spoken of picturing objects.)Since you wish to give a general characterization of truth and ofsemantical rules, you are committed to the claim that in asituation of radical translation it is empirically discoverablewhat the semantical rules concerning picturing are for the nativelanguage in question. That is, you are committed to the claimthat it is empirically discoverable whether or not a given nativeutterance is a correct perceptual response to rabbits, i.e.whether it correctly pictures rabbits. (Quine would of coursedeny this claim. I do not see that you have any adequate reply toQuine -- but let us by-pass that issue; there are enough otherissues.)I must confess that I am still puzzled by something I discussedon pages 20-21 [418?] of my review. Perhaps I can now express mypuzzlement in a clearer fashion than I did there. Given the way Ihave formulated your objection to Tarski (on page one of thisletter), you are committed to characterizing "true-in-L" forvarying L. Since you characterize truth in part in terms ofpicturing, you are committed to characterizing "pictures-in-L"for varying L. Now, in much of what you say, picturing seems tobe something that occurs by way of perceptual responses; indeedone might attempt to formulate a general characterization ofpicturing in terms of perceptual responses and how such responsescan be criticized. But what about Science and MetaphysicsV, 92? There you assert that "singular statements in the languageof microphysics . . . constitute pictures of microphysicalobjects and events . . ." But you also agree that "no singularstatement about microphysical particles can occur in alanguage entry transition or observation" by which I takeit you mean a perceptual response. This seems to break theconnection between pictures and perceptual responses andseemingly permits me to return your complaint against Tarski: youdo not characterize "picturing-in-L" for varying L, since you donot say what picturing is if it is not tied to perceptualresponses. You have not explained what "picturing-in-L" comes toif L is taken to the language of microphysics. That means youhave not explained what "truth-in-L" comes to if L is thelanguage of microphysics.Finally I would like to remark on your discussion of the schema:"p" is true if and only if p. On page 32 you allow a derivationto use a principle of "pragmatic implication" and yourefer me to Moore's discussion of such implications. As I recall,one case Moore discussed involved belief. There is a pragmaticimplication in either direction between "I believe that p" and"p". Notice, however, that it is not true that I believe that pif and only if p. Pragmatic implications do not in generalsupport conditional statements.On pages 34-38 of your letter you discuss the important casewhere "p" takes the form of a basic statement "fa," true if andonly if "f"s picture a. But I do not see that you have shown that"f"s picture a if and only if fa. That ought to follow from youraccount of picturing. I do not see that it does. You haven't saidenough about what picturing is to get that conclusion.I hope that you will think I have learned from your letter. Ilook forward to learning more.Sincerely,Gilbert H. Harman[Contents]FACULTY OF ARTS AND SCIENCESUNIVERSITY OF PITTSBURGHPITTSBURGH, PENNSYLVANIA 15213DEPARTMENT OF PHILOSOPHYNovember 20, 1970Professor Gilbert HarmanDepartment of PhilosophyPrinceton UniversityPrinceton, New JerseyDear Gil:It is about time I put down on paper some of the things Isaid in response to your letter of March 24 during lunch at thatunfindable Italian restaurant. I made notes shortly afterward andhave been writing in fits and starts a formal 'reply' a laSellars-Chisholm. But your letter raises so many importantquestions which I must think through again, that if I wait untilI have them all in hand, I will have written Son of Scienceand Metaphysics. What seems to be called for, therefore,is a series of less ambitious letters which take up separate questions as my ideas about them take shape.First a note about the way in which you relate my distinctionbetween analyzing the sense of 'true' and giving arecursive specification of truth conditions, to Quine'sdistinction between giving a definition of 'analytic-in-L' forvariable L, and giving a definition of 'analytic-in-L0.' Yousuggest that "the semantic theory of truth never defines'true-in-L' for variable L; at best it defines 'true-in-L0' for aparticular language L0." The analogy, as you see it, yields apossible interpretation of my claim that "Tarski's account movesin a large circle," which puzzled you "since Tarski's accountseems to be no more circular than any other recursivedefinition." You find it less puzzling if my claim is interpretedas the "point, not that Tarski's account is circular as anaccount of truth-in-L0, but that it becomes circular if turnedinto an account of truth-in-L for variable L."I think that this is a very helpful move, particularly sinceI independently made closely related distinctions in my essay inthe Carnap volume. Yet I would want to press the point a bitfurther, and distinguish, as I did in that essay, between, on theone hand, 'defining' (to take a simple case)Sentence (in L0)by a recursive specification of sentential expressions of L0, -- which 'definition' amounts to a recursive listing ofsentences of L0, and does not mention (though it makes a tacituse of) the function by performing which these expressionsare sentences of L0 -- and, on the other hand, giving adefinition of 'sentence (in L0)' which both specifies the items in L0, which do the sentence jobandexplicates what the sentence job is.Only the latter would I count as a definition proper of'sentence (in L0).' On the other hand, as I pointed out in theessay, we can introduce 'sentence-in-L0' as a contrived predicatewhich stands to 'sentence (in L0)' as the contrived predicatex and y are the-twins-in-the-Jones-family(where this is introduced to mean x = Tom and y = Dick or x = Jack and y = Jill or x = Bob and y =Carol)stands to x and y are the twins in the Jones family (where this means something likex = Tom and y = Dick and x is connate dual sibling of y orx = Jack and y = Jill and x is connate dual sibling of y orx = Bob and y = Carol and x is connate dual sibling of y)Notice that the former predicate is introduced, and asintroduced does not contain as an element in its sense theconcept of twin. It simply abbreviates the listing oftwins in the Jones family --even though the concept of twin (and, for that matter, of theJones family) is used in picking out the items to be listed.Nevertheless if the stipulation 'catches on', the definition ofthe contrived predicate by means of the disjunction ofconjunctions which follows it would be a definition proper, sincethe predicate, thus introduced, would have this sense.If now, we were to use the definiens of the contrived predicateas the 'definition' of open sentence 'x and y are the twins inthe Jones family, which does contain as an element of itssense the concept of twin, then, in my terminology,we would be giving an extensional definition' of this predicate,but not a definition proper.I have no objection to 'extensional definitions' nor, inparticular, to 'extensional definitions' which list recursively.I also have no objection to such contrived predicates as'sentence-in-L0' (as contrasted with 'sentence (in L0).'Correctly used, both are legitimate and can be clarifying. I do however, think that when their role is misunderstood, they can bethe source of philosophical perplexity, for they may make it seemto philosophers that certain concepts have beenexplicated, when what has been explicated is theircontrived extensional counterparts.Thus even though the contrived predicate 'sentence-in-L0'can be explicated without any explication or the concept ofsentence, the use of this contrived predicate presupposesthe concept of sentence. Again, the explicans ofthe contrived predicate does not explicate even the concept of'sentence (in L0)' -- let alone 'sentence (in L)' for variable L-- the corresponding non-contrived predicate, but simply givesits extensional exposition.Thus, from my point or view, the interesting question aboutTarski's definition of 'true-in-L,' is the following: How are weto understand such statements (at the basic recursive level) asa satisfies 'fx' (in L0)'fx' (in L0) is true of aa belongs to the denotation of 'fx' (in L0).There are two alternatives: (1) 'satisfies' or 'true of' is usedin its ordinary sense, and this ordinary sense involvessomething like my concept of semantic assertibility orcorrectness of dequoting, in which case the account of truth(in L0) brings in concepts of meaning and truth pertaining to ourbackground language, which concepts it does not attempt toexplicate, though it uses them in the process of giving arecursive account of necessary and sufficient conditions forbeing a true sentence (of L0).The second alternative, (2), is to construe x satisfies yas used in the 'definition' of 'true sentence (of L0)' as simplya more complicated case of the type of recursive specificationillustrated byx denotes y =x = 'New York' and y = New York orx = 'Chicago' and y = Chicago orx = 'Pittsburgh' and y = Pittsburgh. . .Here, however, there are two sub-alternatives: (2a) Our accountof 'true (in L,)' doesn't make direct use of the concepttruth-of, the latter coming into the account only as theordinary sense of 'sentence' came into the recursivespecification of sentence (in L0). Or, (2b), the ordinary senseor 'true of' simply is a recursive specification (in ourbackground language) of pairs of expressions and objects. Thelatter, (2b) would amount to the claim that our backgroundsemantical concepts can be explicated in terms of thepattern (where 'Sigma' is a semantical predicate),Sigma (xy) =x = expression-1 and y = item-1 orx = expression-1 and y = item-2 or................................................x = expression-2 and y = item-1 orx = expression-2 and y = item-2 or................................................x = expression-n and y = item-n or................................................................................................Only on alternatives (1) and (2a) would Tarski's account besubject to my charge of 'circularity' as an explication of theconcept of truth. On alternative (2b), the charge of circularitywould fail, because the account of 'true-in-L0' would not makeuse of unexplicated semantical concepts in the backgroundlanguage, because these semantical concepts themselves would havebeen successfully explicated in terms of the resources of thefunctional calculus.Now Tarski seems to claim as philosopher, that hehas explicated the concept of truth (in L0) interms of purely logical resources. From my point or view, thiswould have to mean that his account is circular (i.e. presupposesand does not explicate the concept 'true (in L)' for variable L),unless he can substantiate the thesis that the background conceptof satisfaction or truth of, or belonging to thedenotation of which are used at the ground floor of therecursive account are themselves explicable in terms ofthe resources of the functional calculus.On the other hand, if without advancing any such thesisabout the background concepts, he is simply using thesebackground concepts to give a (recursive) extensionalexposition of 'true sentence (of L0)' or, alternatively, a(recursive) explication of the contrived extensionalpredicate 'true-sentence-of-L0', then he is indeed not guiltyof circularity, for he has simply abandoned the philosophicalclaim to be explicating the intention of 'truth (inL0)', let alone 'truth (in L)' for variable L.Now I am not clear where you stand with respect to all this (given that you find what I am saying intelligible). On the one hand you seem to agree with me that (and I apply your remarks in the second paragraph of your letter to the case of truth) that "any adequate account of [truth] in general would have to refer to 'mental or behavioral or cultural factors' (Quine)" and applaud my attempts to do so in terms of rule governed linguistic behavior (page 2, paragraph 2 ff.). On the other hand, in the paragraph which begins at the bottom or page 1, you seen to go along with the idea that our background concept of satisfaction can be explicated in terms of the patternSigma (x,y) = x = expression-1 and y = item-1 or..............................................................................................................If you can make sense of this, perhaps you can, as usual, helpme say it better. In any event, the above remarks set the stagefor the main purpose of this letter, which is to agree with youthat the account which I give in my original letter of a'pragmatic' sense in which(1) Carnap is wiseimplies(2) It is true that Carnap is wisedoes not explicate the proper sense in which (1) implies (2) and,for that matter (2) implies (1).In other words, I agree that the account in my letter is incomplete and, which is worse, directs attention away from the correct account of the connection between the meaning of truth and the criteria of truth -- in the sense of truth conditions, not evidence. Since I think that I am clear in my own mind about this, I am chagrined that my letter messed this up.If letters could have second editions, I would replace the material from p. 31, second paragraph, to p. 34, paragraph ending at the top of the page, by the following text.[editor's note: These passages have been marked.] The deleted material with an appropriate introduction, could then be added as a relatively uninteresting appendix.The problem is most pressing with respect to basic sentences.Yet to fix our ideas (as Nagel says) it is worthwhile to look ata less problematic case. In what sense does 'p or q' imply 'it istrue that p or q'? Or, to translate this in terms of my analysis,but omitting dot-quote niceties, in what sense does 'p or q'imply ' 'p or q' is semantically assertible (with respect torelevant rules)'?The structure of the answer is given by the recursivespecification of truth conditions. In this respect my accountdoes not differ from that of "Carnap-Tarski semantics." Withrespect to the sense of the answer, however, there is adifference which, though it may seem minor at this stage in therecursive structure, is of a piece with my account of thesense in which basic statements correspond to the world,and my critique of Tarski's explication of this correspondence inhis classic paper on "The Semantic Conception of Truth."As far as structure, then, is concerned, consider the followingschema (where 'p' and 'q' represent basic sentences in thelanguage we use, and ''p'' and ''q'' represent designations ofthese sentences in the metalinguistic level of this language {3}):1. p --> 'p' is true Truth Cond.: basic sentences.2. q --> 'q' is true Truth Cond.: basic sentences.3. p or q --> 'p' is true or 'q' is true (1), (2): Prop.Calc.4. 'p' is true or 'q' is true --> 'p or q' is true Truth Cond. 'or'5. p or q --> 'p or q' is true (3), (4),: Tans. -->.This derivation has two distinctive features. The first is thatpremises (1) and (2) have the same form as the conclusion to bederived, i.e. the form. . . --> '. . .' is trueThis feature is not surprising, for it simply illustrates thefact that we are dealing with a recursive specification of truthconditions (conditions of semantic assertibility). The secondfeature, however, is of greater interest, for it involves thedistinction between semantic assertibility (truth) and acriterion or a ground of such assertibility. Thusstep (4) is an appeal to the principle that a disjunctivestatement is semantically assertible iff at least one ofits disjunct is semantically assertible. And it is of the utmostimportance to realize that in the equivalence'p or q' is SA <--> 'p' is SA or 'q' is SAthe left hand and right hand sides do not have the samesense. For the LHS authorizes the inscription of 'p orq's, whereas the RHS is a disjunction of an authorization toinscribe 'p's and authorization to inscribe 'q's.A possible source of confusion arises most clearly in the caseof conjunction. Thus consider'p and q' is SA <--> 'p' is SA and 'q' is SAHere, again, the equivalence is not an identity of sense, for theLHS authorizes the inscription of 'p and q's whereas the RHS is aconjunction of an authorization to inscribe 'p's and is anauthorization to inscribe 'q's. But is there not also the rule ofConjunction IntroductionFrom 'p' and 'q' to infer 'p and q'according to which, if we are entitled to inscribe a 'p' andentitled to inscribe a 'q' we are entitled to inscribe a 'p andq'? The RHS of the above equivalence authorizes us to inscribe asfollowsp,qp,q. . . .. . . .. . . .. . . .Conjunction Introduction authorizes us to add, in each case, a 'pand q,' thusp,q so, p and qp,q so, p and q . . . .. . . .. . . .. . . .Consequently, it is a fact of the meaning of logical words thatthe authorizations to inscribe 'p's and 'q's carry with them "asa matter of sense" an authorization to inscribe 'p andq's. This is quite true. It does not follow, however, that theauthorization to inscribe 'p and q's consists in aconjunction of an authorization to inscribe 'p's and anauthorization to inscribe 'q's. The situation is rather thatConj. Int. (and its converse Conj. Elim.) give the senseof 'and,' and exhibit conjunction as a purely 'linguistic' orsyntactical concept. Parallel points can be made aboutdisjunction and negation. (Of course, a careful discussion of thesemantics and syntactics of connectives would have to take intoaccount the possibility of non-classical connectives.)Before turning our attention to the distinction between the sense of 'true' as applied to basic sentences and theconditions of its applicability, we should note that theabove derivation ofp or q --> 'p or q' is truecan be matched with a derivation of the converse hypothetical,'p or q' is true --> p or qit goes as follows:1. 'p or q' is true --> 'p' is true or 'q' is true Truth Cond.: 'or'2. 'p' is true --> p Truth Cond.: basic sent.3. 'q' is true --> q Truth Cond.: basic sent.4. 'p' or 'q' is true --> p or q (2), (3), Prop. Calc.5. 'p or q' is true --> p or q (1), (4), Trans. -->.Steps (2) and (3) remain, of course, to be discussed, as do steps(1) and (2) in the converse derivation. My present purpose,however, is to emphasize that the fact that this derivationestablishes that'p or q' is true --> p or qshould not tempt us to overlook the special connectionbetween ''p or q' is true' and 'p or q' which is not a matter ofimplication, and which pertains to the sense of 'true.'Thus, although the sequence'p or q' is trueso, p or qcan be viewed as an enthemyme with the suppressed premise'p or q' is true --> p or qor as an inference licensed by the principleFrom 'p or q' is true' infer 'p or q'which the above derivation would justify, it can also be viewednot as an inference, but as a sequence in which thestatement that a certain kind of action is correct is followed bya performance of that kind of action. The latter way ofviewing the sequence takes into account the sense of thepredicate 'true,' i.e. (on my analysis) 'semanticallyassertible.'The distinction between the truth move and the relatedinference is bound up with the distinction, stressedabove, between the meaning of 'true' ('semanticallyassertible') and the recursive specification of truthconditions. It is because, as I see it, the truth conditionsfor basic sentences consist in matter-of-factual relationsbetween linguistic and non-linguistic items, by virtue of whichthe former picture the latter, that my account of truth breaksout of the family of semantical concepts (true, true of,satisfies, belongs to the extension of, etc.) and brings inpragmatic considerations pertaining to the causal involvement oflanguage users in the natural order.Two points should be carefully noted: (1) By 'relations' in'matter-of-factual relations' I mean relations. That is tosay, I regard the distinction between relations andconnectives as of the utmost importance to soundphilosophy. Thus, when I deny that 'means' stands for a relation,it is because I analyze'dreieckig' (in G) means triangularasThe 'dreieckig' (in G) is a .triangular.i.e. as'dreieckig's (in G) are .triangular.sObviously it would be obtuse to treat 'are' as a relation word orpredicate. A similar point is to be made with respect to 'standsfor' as in'dreieckig' (in G) stands for triangularity.This I 'regiment' by the following steps The 'dreieckig' (in G) stands for triangularity'dreieckig's (in G) stand for triangularity'dreieckig's (in G) stand for the .triangular.'dreieckig's (in G) are .triangular.sWith 'denotes' the same strategy yields the sequenceThe 'Vernunftiges Tier' (In G) denotes Featherless Biped(E s) 'VT' (in G) stands for s, and s is materiallyequivalent to .FB.Where 's' is a predicate variable taking dot-quoted expressionsas its primary substituends. Regimented as above, this becomes(E s) 'VT' (in G) C s and s ME .FB.which is verified by'VT' (in G) C .RA. and .RA. ME.FB.of which the right hand conjunct tells us (in accordance with myconvention of using '.IND [epsilon] RA.' as apredicate which applies to tokens consisting of an INDconcatenated with an .[epsilon] RA.)(IND) .IND [epsilon] RA. is true iff.IND [epsilon] FB. is truewhich, in turn, is equivalent{4} to(x) x [epsilon] RA --> x [epsilon] FB.The second point to be noted, (2), is that if by relations I meanrelations, it should be clear that bysentence affirming a matter-of-factual relationI don't mean simplyContingent multi-term sentence.Thus,Man C Featherless Bipedis a contingent sentence involving the two terms 'Man' and'Featherless Biped,' but it does not, in my terminology, affirm amatter-of-factual relation between men and featherlessbipeds. By parity of reasoning, the facts that'dreieckig' (in G) means triangularand'Vernunftiges Tier' (in G) denotes Featherless Bipedare contingent truths should not lead us to say that theyassert the obtaining of matter-of-factual relations. This letter is designed to lay the foundation for a discussion ofpicturing. Thus I hope to take up in my next [letter] yourquestion concerning 'pictures-in-L' for variable L. I must alsosatisfy you that it follows from my account of picturing that(leaving dot-quote niceties and tenses aside)"a"s which are concatenated with an "f" [f*"a"s] picture a(correctly) iff fa(not, as you put it,"f"s picture a iff fa).If I can pull this off by our satisfaction, I will take a vacation.As ever,Wilfrid Sellars[Contents] PRINCETON UNIVERSITYDEPARTMENT OF PHILOSOPHY1879 HALLPRINCETON, NEW JERSEY 08540December 9, 1970Professor Wilfrid SellarsDepartment of PhilosophyUniversity of PittsburghPittsburgh, Pennsylvania 15213Dear Wilfrid:Three comments on your letter of November 20.(1) According to Tarski, a general notion of truth applicable toany language leads to paradox. So he defined only truth inL0 for a particular language L0. One might extendhis account to any language L' translatable into L0by counting a sentence in L' true just in case itstranslation is true in L0. It is this extension that Isupposed you would find circular. It makes use of a notion oftranslation; and I take it that you think the notion of truthfigures in the explanation of translation.(2) What justifies your assumption that "p or q" isSA if and only if "p" is SA or "q" is SA? Tarskican make use of an analogous assumption because in his definitionof satisfaction he has a clause saying that a sequence satisfiesa disjunction just in case it satisfies at least one disjunct.But your truth definition has no such clause.This is no mere technical problem. For mathematical statementsyou equate semantic assertability with provability. But it is notin general true that a disjunction is provable if and only if oneof its disjuncts is provable.(3) A similar question can be raised about the justification ofthe truth condition cited for "and": "p and q"is SA if and only if "p" is SA and "q" is SA. Inyour letter you find a connection between this truth conditionand the rules of conjunction introduction and conjunctionelimination. Elsewhere, in your discussion of induction, you haverejected the rule of conjunction introduction. Does that mean youalso reject the truth condition "and"?Sincerely,Gilbert H. Harman[Contents]Notes{1}Donald Davidson, "True to the Facts," The Journal ofPhilosophy, 66, 1969, pp. 748-64.[Back]{2}T is the condition that "ones theory of truth should entail, forevery sentence of the language in question an instance of the . . .formulax is true iff pwith a name of the sentence replacing x and with atranslation of that sentence replacing p." (p.13)[412-13][Back]{3} Notice that by not taking into account the internalstructureof 'p' and 'q,' I am abstracting from essential features of thespecification of truth conditions for basic sentences. Thesefeatures concern the subject-predicate structure of basicsentences, to which reference must be made in characterizing thepicturing relationship which defines the truth conditions forsuch sentences. [Back]{4} I realize, of course, that this bald assertion ofequivalenceraises all the issues involved in the controversy over'substitutional' versus 'objectual' quantification. [Back]{*} edited in hypertext by Andrew ChruckyEditor's note: The correspondence, which I probablyobtained from either Sellars or Harman -- though I don't rememberfrom which one of them, consists of five items: (1) a 20 page (double-spaced) typescript of Harman's essay,"Sellars' Semantics," (subsequently published in PhilosophicalReview 79 (1970) 404-419), which is a critical review ofSellars' Science and Metaphysics: Variations on KantianThemes (New York: Humanities Press, 1968); (2) Sellars' letter to Harman, which is a 40 page (double-spaced)critique of Harman's essay; (3) a 3 page (single-spaced) reply by Harman;(4) an 11 page (double-spaced) rejoinder by Sellars; and (5) a final 1 page (single spaced) retort by Harman.Within my replication of the correspondence, next to thereferences to Harman's 20 page typescript of his essay (and someother places), I have inserted in square brackets thecorresponding pages of the published version. Because of the limitations of hypertext, substitutions ofcharacters and designs have to be used. For Greek characters Isimply use their names. The summation sign expressed by an uppercase sigma is rendered by the word "Sigma", while the epsilon ofset membership is expressed by "[epsilon]". The existentialquantifier (inverted E) is rendered by a bold upper case "E";material implication is expressed by a bold "-->"; set inclusion, by abold upper case "C"; and where "L" has the subscript "0", this isrendered as "L0"; the triple bar of material equivalence isrendered by a bold "<-->"; and dot quotation is expressed bybold periods. All underlining and names of books and journalshave been italicized. Minor grammatical and spelling mistakeshave been corrected.I wish to thank Professor Harman for his permission to let mepublish his correspondence on the Internet.[Back][Contents][Return to Main Page] |
|
