Mind Association

On Concept and Object Author(s): Gottlob Frege, P. T. Geach, Max Black Source: Mind, New Series, Vol. 60, No. 238 (Apr., 1951), pp. 168-180 Published by: Oxford University Press on behalf of the Mind Association Stable URL: http://www.jstor.org/stable/2251430 Accessed: 30/08/2010 21:11
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=oup. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Oxford University Press and Mind Association are collaborating with JSTOR to digitize, preserve and extend access to Mind.

http://www.jstor.org

II.-ON

CONCEPT

AND OBJECT

BY GOTTLOBFREGE
fi4r wissenschaftliche (First published in the Vierteljahrsschrift 16 (1892): 192-205.) Philosophie, by P. T. Translation
GEACH.

(Revisedby MAXBLACK.)

and its on intuition in thisQuarterly 192] In a seriesofarticles to BennoKerryhas severaltimesreferred psychicalelaboration, and otherworksof mine,someder Arithmetik my Grundlagen withme. I cannotbut disagreein,g and sometimes timesagreeing be pleased at this, and I thinkthe best way I can show my ofthepointshe contests. is to take up thediscussion appreciation because his opposition This seemsto me all the morenecessary, be whichmight is at least partlybased on a misunderstanding, of whatI say about the concept; and because, sharedby others, is important and thematter thisspecialoccasion, apartfrom even, than seemedto treatment enoughfora morethorough difficult me suitablein my Grundlagen. The word' concept' is used in variousways; its senseis someperhapsa logical,and sometimes sometime, timespsychological, of both. Sincethislicenceexists,it is natural mixture confused that whenonce a usage is adopted it it by requiring to restrict to a shall be maintained. What I decidedwas to keep strictly thisor thatuse is more whetber logicaluse; the question purely is one that I shouldlike to leave on one side, as of appropriate will aboutthemodeofexpression minor importance. Agreement that thereis someeasily be reachedwhenonce it is recognized a specialterm. that deserves thing from his results It seemsto me that Kerry'smisunderstanding his own usage of the word 'concept' confusing unintentionally forwhich withmine. This readilygives rise to contradictions, myusage is not to blame. of 'concept'. 193] Kerrycontestswhat he calls my definition in the first place, that my explanationis not I would remark, meant as a properdefinition. One cannot requirethat everyany more than one can requirethat a thingshall be defined, cannot substance. Whatis simple every decompose chemist sha,ll and whatis logically simplecannothave a proper be decomposed, logicallysimpleis no moregivenus definition.Now something
168

GOTTLOB

FREGE:

ON CONCEPT

AND

OBJECT

169

elements are; it is reached thanmostofthechemical at theoutset has been diswork. If something only by means of scientific coveredthat is simple,or at least must countas simpleforthe forit, sincelanguagewill we shallhave to coina term timebeing, that exactlyanswers. On' containan expression not originally a definilogically simple, ofa nameforsomething theintroduction forit but to lead the reader tionis notpossible; thereis nothing the words as is or hearer,by means of hints,to understand intended. betweenconcept Kerrywould like to say that the distinction passage.",he says,"I and objectis notabsolute. "In a previous the opinionthat the relationbetweenthe have myself expressed is, in a certain contentof the concept and the concept-object a peculiarand irreducible one; but this was in no way respect, ofbeinga concept and boundup withthe viewthattheproperties of being an object are mutuallyexclusive. The latterview no than it would follow, if,e.g., the moie followsfromthe fQrmer relationof fatherand son were one that could not be further reduced,that a man could not be at once a fatherand a son ofthe man whosesorn he was)." (thoughof coursenot e.g. father orhad been,beings on thissimile! If there were, Let us fasten but could not be sons, such beingswould obthat werefathers all men, whoare sons. Now in kindfrom viously be quitedifferent like this that happenshere. The concept(as I it is something hand,a name theword)is predicative.' On theother understand of an object, a proper name, is quite incapable of beingused needs elucidation, as a grammatical predicate. This admittedly it mightappear false. Surely one can just as well otherwise the Great,or is the number that it is Alexander assertofa thing ? four,or is the planetVenus,as that it is greenor is a maimmal the usages this,he is not distinguishing thinks 194] If anybody ofthe word'is '. In thelast two examplesit servesas a copula, as a mereverbalsignofpredication. (In thissense [the German be replacedbythe merepersonalsuffix: wordist]can sometimes In sucha case we say cf.diesBlatt istgrin and diesBlattgrimnt.) prediand thegrammatical thatsomething fallsundera concept, threeexamples,on the cate standsforthisconcept. In the first to other hand, 'is ' is used like the 'equals signin arithmetic, ster is expressan equation.2 In the sentence 'The morning
is, in fact, thereference ofa grammatical predicate. J use theword'equal' and the symbol' = ' in the sense 'the same as', ' no otherthan', 'identical with'. Cf. E. Schroeder, Vorlesungen ueberdie Algebra derLogilc(Leipzig,1890), vol. 1, ?1. Schroeder must however be criticized fornot distinguishing two fundamentally different
2

1 It

12

170

GOTTLOB

FREGE:

Venus', we have twoproper names,'m orning star 'and 'Venus', for the same object. In the sentence' the morning star is a planet'we have a proper name,' themorning star', and a conceptword,'planet'. So faras languagegoes,no morehas happened than that ' Venus' has been replacedby ' a planet'; but really therelation has becomewhollydifferent.An equationis reversible; an object'sfalling undera concept is an irreversiblerelation. In thesentence 'the morning staris Venus ', 'is 'is obviously not themerecopula; its content is an essential partofthe predicate, so that the word 'Venus' does.not constitute the wholeof the predicate.' One mightsay instead: 'the morning star is no otherthan Venus'; what was previously implicitin the single word 'is' is hereset forth in foarseparatewords,and in 'is no otherthan' the word'is' now reallyis the merecopula. What is predicatedhere is thus not Venus but no other than Venus. These wordsstand fora concept; admittedly only one object fallsunderthis,but such a conceptmust still always be distinguishedfromthe object.2 We have here a word 'Venus' that can neverbe a properpredicate, although it can 3 of thiswordis 195] form part of a predicate. IThereference thussomething that can neveroccuras a concept, but onlyas an object. Kerrytoo wouldprobably notwishto disputethatthere is something of this kind. But this would mean admittinga which it is veryimportant to recognize, distinction, what betweeia caD occuronlyas an object,and everything else. And this disevenifit weretrue,as Kerrythinks wouldnotbe effaced tinction it is, that thereare concepts that can also be objects. Thereare,indeed,casesthatseemto support his view. I myself have indicated(in Grundlagen, ?53, ad fin.) that a conceptmay fall undera higherconcept-which,however, must not be conto another. Kerry fusedwith one concept'sbeingsubordinate does notappeal to this; instead, he givesthefollowing example: " the concept'horse ' is a concepteasilyattained", and thinks that the concept'horse ' is an object,in fact one of the objects thatfallunderthe concept' concepteasilyattained'. Quiteso; the threewords" the concept'horse' " do designatean object,, but on that veryaccounttheydo not designatea concept,as I
and the of an object to a conceptit fallsunder, relations; the relation on the Vollwurzel of one concept to another. His remarks subordination 4 does not simply symbol are likewise open to objection. Schroeder's take the place ofthe copula. 2 Ibid., ?51. 1 Cf.myGrundlagen, ?66,footnote. 3 Cf. mypaper 'On Sense and Reference'(Ueber Sinn undBedeutung), furPhil. undphil.Kritik. shortly to appearin the Zeitschrift

ON CONCEPT AND OBJECT

171

1 am using the word. This is in fullaccordwiththe criterion alwaysindicates an object, definite article gave-that thesingular a concept-word.1 articleaccompanies whereas the indefinite Kerryholds that no logical rules can be based on linguistic that distinctions;but my own way of doingthis is something nobodycan avoid wholays downsuchrulesat all; forwe cannot apartfrom withone another cometo an understanding language, and so in the end we mustalways relyon otherpeople's underand sentence-construction in essenstandingwords,inflexions, I was not tiallythe same way as ourselves. As I said before, to give a definition, but only hints; and to this end I' trying forthe Germanlanguage. It is appealed to the generalfeeling hereverymuchto myadvantagethat thereis such good accord and the real one. As regards distinction betweenthe linguistic to our rule no exceptions articlethereare probably the indefinite like ' Ein apart fromobsolete formulas at all forus to remark, is not so simpleforthe edlerRath' [' Councillor']. The matter in the plural; but definite article,especially does not relate to this case. In the 196] then my criterion onlywhena singular, so faras I can see, the matteris doubtful 'the Turk singular takes the place of a plural,as in the sentence animal'. These besiegedVienna', 'the horse is a four-legged as special ones that the value of cases are so easily recognizable our ruleis hardlyimpairedby theiroccurrence. It is clearthat sentence'the Turk' is the propername of a people. in the first a The second sentenceis probablybest regardedas expressing animals' or judgment, say 'all horsesare four-legged universal animals'; these 'all properly horsesare four-legged constituted unsuitable; will be discussedlater.2 Kerry calls my criterion
on p. 80. ?51; ?66,footnote; ?68,footnote Grundlagen, the scope of the stateexaggerate equivalent, expressions are nevercompletely mentthatdifferent linguistic language. One intoanother that a wordcan neverbe exactlytranslated and say thatthesamewordis nevertaken might perhaps go evenfurther, in quite the same way even by men who sharea language. I will not enquireas to the measureof truthin these statements; I would only different expressions quite oftenhave someemphasize that nevertheless in common, I callthesense, which or,in thespecialcase ofsentences, thing thatthesame words; we mustnotfailto recognize thethought. In other maybe variously expressed; thusthe difference sense, thesamethought, or shading, does not hereconcern the sense,but onlythe apprehension, forlogic. It is possible forone ofthethought, and is irrelevant colouring thananother; and, for sentence to giveno moie and no less information has a common stockofthoughts. mankind all themultiplicity oflanguages, on theplea thatthis wereforbidden If all transformation oftheexpression be crippled; forthe as well,logicwouldsimply wouldalterthe content
2 Nowadayspeopleseeminclined to

172

GOTTLOB

FREGE:

' the conceptthat I am now he says, in the seintence forsurely, ofthe concept' thenamecomposed about is an individual talking eightwordsstandsfora concept; but he is not takingthe first laid down word' concept'in mysense,and it is notin whatI haive lies. But nobody can requirethat my that the contradiction shall agreewithKerry's. modeof expression by an thatherewe are confronted It mustindeedbe recognized whichI admitcannotbe avoided,ifwe oflanguage, awkwardness e.g.,the is not a concept,'whereas, say that the concepthorse is a cityand thevolcanoVesuviusis a volcano. 197] cityofBerlin the departure that justifies Language is here in a predicament of our case is indicatedby Kerry custom. The peculiarity from around' horse'; I use by meansofthe quotation-marks himself, italicsto the same end. Therewas no reasonto markout the words 'Berlin' and 'Vesuvius' in a similarway. In logical about a discussionsone quite oftenneeds to assert something usual forsuchassertion', thisin the form and to express concept, viz., to make what is assertedof the conceptintothe content one would expect of the grammatical predicate. Consequently, of the grammatical subjectwould be the conthat the reference cept; butthe conceptas suchcannotplaythispart,in viewofits into an object,2 be converted nature; it must first predicative moreprecisely, by an object. We desigrepresented or,speaking the words'the concept'; e.g. nate this object by prefixing 'The conceptman is not empty'. name,3 as a proper threewordsare to be regarded Here the first than ' Berlin' or which can no more be used predicatively ' Vesuvius '. Whenwe say ' Jesusfallsunderthe conceptman', is: aside the copula,the predicate then,setting ' someonefalling underthe conceptman' and thismeansthe same as: a man
the to recognize without trying task of logic can hardlybe performed wouldthenhave all definitions guises. Moreover, in its manifold thought as false. to be rejected 'this rose 1A similar thesentence happenswhenwe sayas regards thing 'is red' belongsto the subject'this predicate is red': the grammatical ' is red " are not a predicate rose Here the words" The grammatical calling buta subject. By theveryact ofexplicitly predicate grammatical it ofthisproperty. we deprive it a predicate, p. X. 2 Cf. myGrundlatgen, 3 I call anything nameifit is a signforan object. a proper

ON CONCEPT

AND

OBJECT

173

But the phrase 'the conceptman' is onlypartofthispredicate. Somebody natureofthe might urge,as againstthe predicative we speak of a subject-concept. But concept, that nevertheless evenin suchcases, e.g.,in the sentence ' all mammalshave red blood' we cannot fail to recognizethe predicativenature1 of the concept; forwe could say instead: 198] 'whateveris a mammalhas red blood' ' if anything or: is a mammal, thenit has red blood'. der Arithimetik, When I wrotemy Grundlagen I had not yet 2 and so, made the distinction betweensense and reference; ' conitent under the expression of a possible judgment', I was combiningwhat I now designate by the distinctivewords I no longeren'thought' and 'truth-value'. Consequently, I thengave (op. cit.,p. 77), as tirelyapproveof the explanation regardsits wording; my view is, however, still essentially the same. We may say in brief, taking'subject' and 'predicate' in thelinguistic sense: A concept is thereference ofa predicate; An object is something that can neverbe the whole reference of a ofa subject. It mustherebe predicate, but can be the reference remarked thatthewords' all ', ' any ', ' no ', ' some ', are prefixed to concept-words.In universaland particularaffirmative and negative we are expressing relations sentences, between concepts; we use thesewordsto indicatethe specialkindofrelation. They are thus,logically speaking, notto be moreclosely associatedwith the concept-words that follow them,but are to be relatedto the sentence as a whole. It is easyto see thisin thecase ofnegation. If in the sentence ' all mammals are land-dwellers' ' expressedthe logical subject of the the phrase 'all mammals predicateare land-dwellers, then in orderto negate the whole
I What I call herethepredicative oftheconcept nature is just a special case oftheneedofsupplementation, the ' unsaturatedness ', thatI gave as theessential feature of a function in mywork Funktion undBegriff (Jena, 1891). It wasthere scarcely possible to avoidtheexpression 'the function F(x) ', although theretoo the difficulty arose that the reference of this expression is nota function. 2 Cf. my essay 'Sense and Reference' in the Zeitschrift far Phil. und phil. Kritik.

174

GOTTLOB

FREGE,

we shouldhave to negatethe predicate: 'are not landsentence dwellers'. Instead, we must put the 'not' in frontof 'all '; from whichit followsthat 'all' logicallybelongswiththe predicate. On the otherhand, we do negate the sentence' The concept mammalis subordinateto the concept land-dweller' by negating the predicate: 'is not subordinate to the concept land-dweller'. If we keep it in mindthat in my way of speakingexpressions like 'the conceptF' designate not concepts but objects,mostof Kerry'sobjections 199] already collapse. If he thinks (cf. p. 281) that I have identified conceptand extension of concept,he is mistaken; I merelyexpressedmy view that in the expression 'the number that applies to the conceptF is the extensionof the concept equunumer,us tg the concept F' the words ' extensionof the concept' couldbe replacedby ' concept'. Noticecarefully that here the word 'concept' is combinedwiththe definite article. Besides thiswas onlya casual remark; I did not base anything upon it. Thus Kerrydoes not succeedin filling the gap between concept and object. Someonemightattempt, to make use of however, in this sense. I have said that to assign a my own statements number involvesan assertionabout a concept; 1 I speak of -properties assertedof a concept, and I allow that a conceptmay fall undera higher one.2 I have calledexistence a property of a concept. How I mean thisto be takenis best made clearby ai, ' there is at leastone squarerootof4 ', example. In thesentence we have an assertion, not about (say) the definite number 2, nor about - 2, butabouta concept, root square of4; viz.,thatit is not empty. But if I expressthe same thought thus: 'The concept square rootof 4 is realized', then the first six wordsformthe propername of an object,and it is about this object that someis asserted. But noticecarefully thing thatwhatis asserted here is not the same thingas was assertedabout the concept. This willbe surprising onlyto somebody whofailsto see thata thought can be splitup in manyways,so thatnowonething, nowanother, appearsas subjector predicate. The thought itself does not yet determine whatis to be regarded as the subject. If we say 'the subjectofthisjudgment',we do not designate anything definite unlessat the same time we indicatea definite kind of analysis; as a rule,we do thisin connexion witha definite wording. But we must neverforget that different sentencesmay expressthe
1 Grundlatgen, ?46.
2

Ibid., ?53,

ON CONCEPT

AND

OBJECT

175

same thought. For example,the thoughtwe are considering 4: could also be taken as an assertion about the number of that there is something 'the number4 has the property whichit is the square'. now one,now another, Languagehas meansofpresenting part is the as thesubject; oneofthemostfamiliar 200] ofthethought distinction ofactiveand passiveforms. It is thusnotimpossible shouldmakeit appear a giventhought thatone way ofanalysing as a particular as a singular judgment; another, judgment; and us a third,as a universaljudgment. lt need not'thensurprise as an assertion thatthe same sentence about a may be conceived about an object; onlywe must conceptand also as an assertion 'there is different.In the sentence thatwhatis asserted observe to replacethe is at least one square root of 4 ' it is impossible words' squarerootof 4 ' by ' the conceptsquareroot of4 '; that is, the assertion that suitsthe conceptdoes not suit the object. does not present our sentence the conceptas a subject, Although as expressing about it; it can be regarded it assertssomething fallsundera higher one.' But this does the factthat a concept between notin any way efface the distinction objectand concept. We see to beginwiththat in the sentence'there is at least one square root of 4 ' the predicativenatureof the conceptis not thathas theproperty of belied; we couldsay ' thereis something 4 when multiplied '. Hence what is givingthe resiult by itself here assertedabout a conceptcan never be assertedabout an name can neverbe a predicative object; fora proper expression, thoughit can be part of one. I do not wantto say it is falseto assertabout an object what is assertedhereabout a concept; I want to say it is impossible, senseless, to do so. The sentence ' is neither 'there is JuliusCaesar true nor false but senseless; ' has a the sentence'there is a man whosenameis JuliusCaasar sense,but hereagain we have a concept, as the indefinite article shows. We getthesamethingin the sentence 'there is onlyone Vienna'. We mustnot let ourselvesbe deceived because language oftenuses the samewordnowas a proper name,now as a concept-word;in our example,the numeralindicatesthat we have the latter; ' Vienna' is herea concept-word, like ' metropolis'. Using it in this sense, we may say: 'Trieste is no Vienna'. If, on the otherhand, we substitute' JuliusCoesar' six wordsof the 201] forthe propername formed by the first
1 In my GrundlagenI called such a concept a second-orderconcept; in my work Funktionund Begriff I called it a second-level concept, as I shall do here.

176

GOTTLOB

FREGE:

sentence'the conceptsquareroot of4 is realized', we get a sentencethat has a sensebutis false; for theassertion thatsomething is realized (as the word is beingtaken here) can be trulymade only about a quite special kind of objects,viz., such as can be designated by proper namesofthe form 'the conceptF '. Thus thewords' the concept squareroot of4 'have an essentially different behaviour, as regardspossiblesubstitutions, from thewords ' square rootof4 ' in our original sentence; thatis, thereference of the two phrasesis essentially different.' Whathas beenshown herein one exampleholdsgoodgenerally; the behaviour oftheconcept is essentially predicative, evenwhere something is beingassertedabout it; consequently it can be replaced thereonlyby another concept, neverby an object. Thus theassertion thatis made abouta concept does notsuitan object. whichconceptsfallunder,are essentially Second-level concepts, different from first-level whichobjectsfallunder. The concepts, relationof an object to a first-level conceptthat it fallsunderis to from the (admittedly different similar)relationof a first-level a second-level concept. (To do justiceat once to the distinction and to the similarity, we mightperhaps say: An object falls a second-level undera first-level concept; a conceptfallswithin of conceptand object thus stillholds, concept.) The distinction withall its sharpness.2 Withthis therehangstogether what I have said (Grundlagen, ' and ' mark'; ?53) about my usage of the words ' property once moreto this. Kerry'sdiscussion givesme occasionto revert The wordsserve to signify in sentenceslike ' 0 is a relations, property ofr and 'O is a markofQ'. In mywayofspeaking, a thingcan be at oncea property but notofthesame and a mark, thing. I call the conceptanderwhichan object fallsits properties; thus 'to be 1 is a property of r'
[When Russell says that expressions like 'the King ofFrance' are not names but incomplete symbols, he is saying what would be put thus in Frege's terminology: " In 'the King of France is bald', 'the King of France' is not a name of an object; what it stands for is somethinginconcept, withinwhich the concept bald complete,ungesdttigt-a second-level is falsely asserted to fall. The second-level concept in question is the falls whois a King ofFranceand somebody under concept: concept which concept falls nobody whom is a King ofFrance; no first-level apartfrom
2

1 Cf.my essay 'Sense and Reference' (cited above).

this,becausenobodyis a King ofFrance." within

Sinn undBedeutung, descriptions.Cf. Ueber accountof definite different pp. 39-42.-P.T.G.]

It should, however, be emphasized that Frege himselfgives an entirely

ON CONCEPT

AND

OBJECT

177 '

way of saying: 1S just another ' r fallsunderthe conceptof a 0, X, and IF, I may combine If the object r has the properties if I saLy that r has the themintoQ; so thatit is the same thing property Q, or, that r 202] has the properties 0, X, and YF. I then call 0, X, and !f of F. marksof the conceptQ, and, at the same time,properties of Z to r and to Q are quitedifferent, It is clearthattherelations different terms are required. r falls and that consequently is itself a concqept, cannotfall underthe concept ; butQ, which concept underthe first-level concept'P; only to a second-level could it stand in a similarrelation. Q is, on the other hand, to i. subordinate an example, Instead of saying: Let us consider '2 is a positivenumber' and '2 is a wholenumber' and '2 is less than 10' we may also say less than 10'. '2 is a positivewholenumber Here to be a positive number, to be a whole number, to be less than10, of the object 2, and also as marksof the appear as properties concept number lessthan10. whole positive nora wholenumber, norless than 10. It This is neither positive, to the conceptwhole but does not is indeed subordinate number, fall underit. Let us now comparewiththis what Kerrysays in his second the resultof article(p. 224). "By the number 4 we understand objecthereoccurring combining 3 and 1. The concept additively is the numericalindividual4; a quite definite numberin the natural number-series.This object obviously bears just the marksthat are named in its concept,and no othersbesidescounting as praprit we refrain, as we surely must,from provided numerous relations to all other of the object its infinitely individualnumbers; ('the ' number4 is likewisethe resultof 3 and 1.)" additivelycombining

178

GOTTLOB

FREGE:

and mark property between We see at oncethatmydistinction the herebetween over. Kerrydistinguishes is herequite slurred that thisdistinc4. I mustconfess 4 and 'the ' number number 4 is to be a concept; to me. The number tionis incomprehensible and none otherthan 'the' number4 is to be a concept-object, 4. It needsno individual thenumerical between thatwhatwe have hereis notmydistinction 203] proof whatwas floating and object. It almostlooksas though concept Kerry'smindweremydistinction before veryobscurely) (though 4 '. ofthewords' the number between thesenseand thereference ofthewordsthatcan be said to be the But it is onlythe reference 3 and 1. combining resultof additively ' the Again,how are we to take the word' is ' in the sentences 3 and 1 ' and number4 is the result of additivelycombining 3 and 1 " ? " ' the ' number combining 4 is theresultofadditively Is it a merecopula,or does it helpto expressa logicalequation? ' result', case, ' the ' wouldhave to be leftout before In the first and the sentences would go like this: 3 and 1'; combining 4 is a resultof additively 'The number 3 and 1." 4 is a resultof additively " ' The' number combining by In that case, the objectsthat Kerrydesignates 4" ' the number 4 ' and " ' the ' number wouldbothfallunderthe concept 3 and 1. combining result ofadditively therewas And thenthe only questionwould be what difference between theseobjects. (I am hereusingthewords'object ' and ' concept' in my accustomedway.) I shouldexpressas follows to say: trying what Kerryis apparently and those alone,whichare 'The number4 has those properties, 3 and 1.' marksofthe concept: result combining ofadditively of our two the senseof the first I shouldthen expressas follows sentences: 'To be a number4 is the same as beinga resultof additive of 3 and 1 '; combination just now to have been Kerry's In that case, what I conjectured could also be put thus: intention and those alone, which 'The number4 has those properties, 4 '. are marksofthe concepta nurmber this is true.) (We need not heredecidewhether

ON CONCEPT

AND

OBJECT

179

articlein the 204] The invertedcommas around the definite 4 " could in that case be omitted. words"'the' number But in theseattempted we have assumedthat interpretations in at least one ofthetwo sentences the definite in front articles of ' result' and ' number 4 ' wereinserted onlyby an oversight. If we take the wordsas they stand, we can only regardthem as havingthe senseofa logicalequation,like: 'The number4 is none other than the resultof additively 3 and 1'. combining The definite articlein front of 'result' is herelogicallyjustified only if it is known(i) that thereis sucha result(ii) thatthereis notmorethanone. In thatcase,thephrasedesignates an object, and is to be regarded as a proper name. If bothofour sentences wereto be regardedas logical equations,then,since theirright it wouldfollow from sides are identical, themthat the number 4 is ' the ' number 4, or,ifyou prefer, thatthenumber 4 is no other than ' the ' number4; and so Kerry'sdistinction would have been proveduntenable. However,it is not my present task to pointout contradictions in his exposition; his way oftakingthe words ' object ' and 'concept' is not properly here. my concern I am onlytrying to set my own usage ofthesewordsin a clearer light,and incidentally showthat in any case it differs from his, whether that is consistent or not. I do notat all disputeKerry'sright to use the words' concept' and ' object' in his own way, if onlyhe wouldrespectmy equal and admitthat withmy use oftermsI have got hold of a right, distinction of the highestimportance. I admit that thereis a quite peculiarobstaclein the way of an understanding withmy reader. By a kind of necessityof language, my expressions, takenliterally, sometimes missmythought; I mention an object, when what I intendis a concept. I fullyrealize that in such cases I was relying upona readerwhowouldbe readyto meetme half-way-whodoes not begrudge a pinchof salt. createddiffiI Somebodymay thinkthat this is an artificially culty; that thereis no need at all to take account of such an unmanageable thingas what I call a concept; that one might, likeKerry, an object'sfalling regard undera concept as a relation, in which thesamething couldoccurnowas object,nowas concept. 205] The words' object' and 'concept ' wouldthenserveonly to indicatethe different in the relation. This may be positions done; but anybody whothinks the difficulty is avoidedthisway is verymuchmistaken; it is onlyshifted.For notall thepartsof a thought can be complete; at least one mustbe 'unsaturated',

180

GOTTLOB FREGE:

ON CONCEPT AND OBJECT

they would not hold together. For or predicative; otherwise example, the sense of the phrase ' the number2 ' does not with that of the expression'the conceptprime hold together number'withouta link. We apply such a linkin the sentence 'the number 2 falls underthe conceptprimenumber'; it is in thewords'falls under', whichneedto be completed contained in two ways-by a subjectand an accusative; and onlybecause as a ' are theycapable ofserving theirsenseis thus ' unsaturated in this twofold link. Only whentheyhave been supplemented respectdo we get a completesense,a thought. I say that such words or phrases stand for a relation. We now get the same to avoid forthe forthe relationthat we weretrying difficulty ofan objectto the concept concept. For the words'the relation it fallsunder' designatenot a relationbut an object; and the number', 2 ','the concept prime names' thenumber threeproper 'the relationof an objectto a conceptit fallsunder', hold aloof twodo by themselves; just as muchas thefirst one another from we get no sentence. It is thus howeverwe put themtogether, the ' unsaturatedfrom arising us to see thatthedifficulty easyfor but not can indeedbe shifted, ness ' of one part of the thought avoided. 'Complete' and 'unsaturated' are of course only of speech; but all that I wishor am able to do hereis to figures give hints. ifthereader It maymakeit easierto cometo an understanding undBegriff.For overthe question comparesmy workFunktion we comeup against in Analysis, whatit is thatis calleda function it willbe found investigation the same obstacle; and on thorough on the natureof our that the obstacleis essential,and founded of language; that we cannot avoid a certaininappropriateness it but to expression; and that thereis nothingfor. linguistic realizethis and alwaystake it intoaccount.