DOI: 10.24308/iass-2014-017

Charles University, Prague, Czech Republic

michal.karla@gmail.com

Abstract

The aim of the proposed paper is twofold: (i) to present a case study of Peirce’s early (i.e. from the period before his 1867 publication of “On a New List of Categories”) notion of interpretant and its origins, and (ii) to give an account of certain general features of Peirce’s semiotic of the period and situating his thought about signs in the context of his metaphysical and logical writings. I would like to show how the conception of inter­pretant stems out of Peirce’s introducing information as a property of terms / propositions / arguments. Infor­mation, according to Peirce, is a product of connotation (comprehension) and denotation (extension), and is represented by an interpretant-sign, being typically the conclusion of an argument, thus representing argu­ment’s premises. I will then proceed in showing the exact role which interpretant together with the notion of information play in Peirce’s early treatment of deduction, induction and hypothesis as different species of the argument-genus, con­centrating mainly on his conceptions and arguments either absent in his later treatments of the same subject, or unintelligible without the knowledge of the earlier theory, e.g. the concept of symbolization and its employment in Peirce’s proofs of the validity of the three species of argument.

In the second part of the paper I will focus on showing certain distinguishing features of Peirce’s early semiotic, building on the concepts examined in the first part. Here I would like to present and argue for a view in which Peirce’s semiotic can be seen as a product and synthesis of his earlier metaphysical and logical ideas. Since several of these connections have been already given account of, I will pay attention mainly to those having direct bearing upon the theory of interpretant and information examined. I consider these being the theory of the three worlds (and most importantly the notion of a formal world), the metaphysical notions of truth, externality and innateness from his Treatise on Metaphysics (W1: 57-84), and the very definition of metaphysics as “analysis of conceptions” (ibid.).

I will use Peirce’s manuscripts of the given period (contained especially in the first volume of Writings) as my primary source. During the course of the paper, I will refer to existing studies more or less concerning the same topic (e.g. Murphey’s The Development of Peirce’s Philosophy, Esposito’s Evolutionary Metaphysics) with some critical remarks.

Subject of the paper is Peirce`s first account of semiotic from 1865 and its explanation in the light of his earlier metaphysical writings. Peirce`s semiotic is understood as a fulfillment of the requirements of metaphysical inquiry being defined as an investigation of the conditions of possibility of empirical science, proceeding by a logical method constituted by unpsychological, formal treatment of thought.

1. Metaphysics and Representation

Peirce understood metaphysics in two ways, as “the philosophy of primal truths” (W1: 59) and “the analysis of conceptions” (W1: 63). The latter`s subject is an explanation of meaning of terms such as “cause, substance, necessity” (W1: 240) and the like, the most fundamental of these conceptions being the categories.[1] We will not follow this part of Peirce`s metaphysics any furhter in this paper.[2] In accordance with the former definition, metaphysics as “the philosophy of primal truths” (W1: 59) is the study of “primary conditions of all science” (W1: 59), its major-premisses (W1: 73), truths being predicated “a priori of the objects of experience” (W1: 61), or warrants “for inferring something not observed” (W1: 152).

What Peirce actually wrote in 1861-2 “Treatise on Metaphysics” and some other subsequent works counts rather as qualified suggestions than a fully developed theory? As I will argue it is Peirce`s semiotic where we can find the actual fulfillment of these suggestions. Before that let us review some basic features of Peirce`s metaphysics.

The center Peirce`s reflections circle arround is a proposition that experience can supply our reasoning with minor premisses only, i.e., particular propositions.[3] No universal proposition can be given in observational experience.[4] If we are thus to ampliate our knowledge beyond the particular facts given in experience, we must assume the truth of a general proposition not derived from experience. Peirce`s approach is transcendental in the sense that it deals with “primary conditions of all science” (W1: 59).[5] It is on the other hand not critical in Kant`s sense,[6] that of reason`s inquiry into its own capacities, for he considers this approach as petitio principii (W1: 73, 77–78; Murphey 1993: 26-7), and the notion of unknowability of the noumenon as self-contradictory (W1: 40, 60). In his view it is objects which are the source and conditions of their knowledge. Peirce so entitles his approach as “uncritical transcendentalism” (W1: 79). The question of such metaphysics is in the first place that of accordance of a representation with its object, or, in Peirce`s words: “How should the conceptions which spring up freely in our minds by virtue of the constitution thereof be true of the outward world?” (W1: 79). The method has to be devoid of any reference to psychology. Psychological (identified by Peirce with Kant`s) method “starts by drawing the conceptions from the system of psychology and reasoning to their logical relations and meaning.” (W1: 63) In contradistinction to that Peirce suggest an unpsychological treatment of metaphysics, which “draws conceptions from no system but from the thoughts as they present themselves in their logical form - examining them logically – and finally puts them in their right place in the mind.” (W1: 63). One can taste here the appetizer of Peirce`s much later Phenomenology or Phaneroscopy as a study of “the collective total of all that is in any way present to the mind” (CP 1.284). More importantly, Peirce is suggesting here a view of logic which is materially indifferent. As a study of form it does not need to consider the matter which embodies that form (the matter of thought) or the way in which this matter is present to us. “[M]eaning resides in words and other material representations” (W1: 306) Quite the contrary Peirce suggests that first we have to understand the concepts and their relations of forms of thought because the mind and its constitution are dependent upon them and not vice versa.

The main question of Peirce`s metaphysics is “How are syntetical judgment in general possible” (W1: 248)? – “how can material inference be valid” (W1: 240) – “what is the ground of scientific presumption and hypothesis” (W1: 248) - What are the grounds of scientific inference? (W1: 279) The answer was given later, when he actually fulfilled the proposed metaphysical approach in his logic.

2. Towards the definition of logic

The purpose of this section is to show how Peirce arrived at his definintion of logic as “objective symbolistic” (W1: 303). In 1865 Peirce gave a series of lectures at Harvard University which he called “On the Logic of Science” (W1: 161). The object of this logic is a study of synthetic, ampliative, or material inference, which is usually not considered to be a part of the logic proper. Although Peirce`s concern is primarily with the scientific, or ampliative, reasoning we will see that in the end he presents a unified theory of all possible reasoning, offering principles for its generic unity and specific difference as well.

The main task of the first Harvard Lecture is to answer the question: What does logic study? What is its subject-matter? Peirce finds out that “The one great source of error in all attempts to make a Logic of Science has been utter misconception of the nature and definition of logic.” (W1: 163) The goal then is first to identify and clear up such misconceptions. So, the multitude of definitions of logic can be, Peirce observes, arranged into two classes: “those which do not and those which do give to logic a psychological or human character.” (W1: 163) Peirce`s position rests in the former approach.

Since Kant, there has been a vast majority of the suffrages of logicians in favor of his definition of logic which is as follows - the science of the necessary laws of the Understanding and Reason - or what is the same thing – the science of the sheer Form of thought in general. Observe the two branches of this statement the former more psychological the latter scarcely at all so; one has two faculties and their capacities, the other thoughts as objects with forms. This is certainly the best definition yet given. (W1: 164)

Peirce thus sets out in the direction of unpsychological view of logic, of the formal treatment of thought. But he does not stop here and suggests that “we ought to adopt a thoroughly unpsychological view of logic.” (W1: 164) This thorough-going antipsychologism rests on the contention that the forms studied by the logician does not have to be forms of thought only (see W1: 164-5). Peirce speaks of either “form” or “logical character” (W1: 164) Every such form or character is said to have its “continual determinator”, which is any kind of matter in which the logical form could be embodied. “[A] continual determinator of a form is that in which the form inheres by the definition of the realition of substantia et accidens.” (W1: 165) So the subject-matter of logic is further approximated: it is any substance (be it mental or material) capable of being a vehicle of – or, which is capable of being predicated to embody – a certain logical character or form. Thus we move from approximating logic as a formal treatment of thought to the understanding of logic as being entirely materially indifferent.

The proposition that there are substances which are continual determinators of forms studied by logic is very soon replaced by an equivalent proposition that logic is a study of representations (W1: 169). Since we can obviously come with various examples of representations which are not logician`s concern, like a portrait, before proceeding further we have to discuss what representation is as a genus and to which species it can be divided in order to find out which species of representation is the subject-matter of logic.

In the period we are concerned with Peirce generally distinguishes three kinds of representations: copy, sign, symbol. They are classified “according to the different ways in which they may accord with its object.” (W1: 169) This accordance is synonymous to truth (W1: 79–80; 169–170). Representations are then divided by a different ways in which they may accord with its object. First kind of representation Peirce calls a copy, which he defines as consisting in the “sameness of predicates” of the representation and its object. From this definition can be inferred that any thing is a copy of some other thing given two limiting cases: if a representation had all the predicates of its object, the two would become identical (granted Leibniz`s principle of the identity of indiscernables), on the other hand “no two things are so different as to resemble each other in no particular” (W1: 170).[7] So for any two things there always is some predicate which they have in common, and some other, which they do not. Therefore, every copy is a partial truth and partial falsehood (W1: 170). Logical representations, on the other hand, “have absolute truth and falsehood” (W1: 170), and copies thus are not logician`s concern. Second of the three species of representation is a sign. It has two distinctive characteristics. It is a representation of an individual object or objects, and its denotation is “fixed upon by convention” (W1: 170). As complete determination and individuality of reference is a limiting and unreachable character of a copy, so does sign lack “genuine generality” (W1: 170) which it cannot attain. It can denote a plurality of objects, but each one of them must be imposed upon the sign. Becase “logic deals only with general terms”, a sign cannot be a proper object of logical study (W1: 170). The conditions which are necessary for a representation`s being logical are met in the symbol. “The third kind of truth or accordance of a representation with its object, is that which inheres in the very nature of the representation whether that nature be original or acquired. Such a representation I name a symbol.” (W1: 170)

In the Harvard Lectures symbol is seen from a certain developmental point of view which is absent in Peirce`s later writings. It is presented as originally being a sign in the above sense, i.e., being a representation of an individual conception. Through the process of its use it became more and more closely associated with that conception as to become in the end identical with it. For example the word “mushroom” was in the beginning an individual sign of the conception of this organism which as a mental conception was general. It signified the idea of a mushroom mediately throught the concept to which it was imposed. In the end the word became so intimately connected with the conception that it started to signify it immediately. “Now this readiness of excitation obviously consists in this; namely, that we do not have to reflect upon the word as a sign but that it comes to affect the intellect as though it had that quality which it connotes. I call this the acquired nature of the word, because it is a power that the word comes to have, and because the word itself without any reflection of ours upon it brings the idea into our minds.” (W1: 172–173)[8]

After this classification is provided, Peirce may return to the question of the subject-matter of logic. “I define logic therefore as the science of the conditions which enables symbols in general to refer to objects” (W1: 175). Logic is conceived of as “objective symbolistic” (W1: 303), that part of semiotic (“the general science of representations”, W1: 174) which deals with symbols, but not from all points of view (since symbol is related to its own essence, its object, and equivalent representation addressed by it), only as considering its objective relation. Thus Peirce`s unpsychological treatment of logic is based upon semiotic. The semiotic does not rest on any reference to the mind, but on the other hand, as far as mind has to be considered at all, it must obey the logical laws, which have a higher authority than the laws of the mind.

3. Objective Symbolistic

Such account would be incomplete, if we did not consider some other notions, which are of fundamental importance for Peirce`s project of offering the grounds of inference. One of these is the notion of interpretant. Interpretant is a representation representing some other representation as representing a certain object and therefore is itself a “more complicated” representation of the same object, being more complicated because it does not represent the object only, but also its representation. (see W1: 303) The example of interpretant par excellence is a conclusion of an argument – it represents the premisses as being a representation of an object (subject-term) as the conclusion itself is. Thus it has a crucial role in distinguishing the kinds of inference – every inference as symbolization (symbol-making) is a production of an interpretant. And the inferences are then distinguished by way of how this interpretant is produced. The quantity of an interpretant Peirce calls information (W1: 465). Every symbol posesses, along with the more familiar connotation and denotation, this quantity. Connotation of a symbol-term is a sum total of marks which are predicated whenever the symbol is predicated of something. Denotation is a sum total of individuals of which a symbol may be trully predicated. Information is defined as the product of connotation and denotation. Two cases can be distinguished. In analytic inference the product of any term`s denotation and connotation is constant in the premisses and the conclusion – the increase of one quantity is always accompanied by the decrease of the other and vice versa – so analytical inference does not bring any new information. In the case of syntehtic inference on the other hand, the increase of one quantity is possible with the other remaining constant – so new information is added either by means of extending denotation (as in induction) or connotation (as in hypothesis) of term in the conclusion as compared to the premisses. While copies have connotation only and sign denotation only, it is and essence of any symbol to have information, that is, to denote by connoting. (W1: 272)

The last among these notions is symbolization. It is a notion identical with inference, they being “the same notions” (W1: 280). Symbolization is making of a symbol-conclusion out of a symbol-premisses, and it also accords with the definition of interpretant – the premisses are symbolized to the conclusion which is their interpretant. Symbolization is conceived as replacing the term in the premisses by either (a) one having the same information, as in deduction, or (b) by a more connotative one, as in hypothesis, or (c) by a more denotative one as in induction. These correspond to the three possible objects or references of the symbol (symbol, form, thing)[9]. The transcendental principle upon which the validity of inferences, their generic unity and specific difference are grounded Peirce thus formulates as “All things, forms, symbols are symbolizable.” (W1: 282; see also W1: 280–281). We may so turn to the examination of grounds of inference.

The core of Peirce`s explanation is that the ground of an inference is valid if it can be derived by the very same inference it is intended to support. This rests upon a premiss that the kinds of reasoning are mutually irreducible (which itself rests upon the fact that they exhibit the forms of the three syllogistic figures which Peirce demonstrated to be irreducible[10]). For a weaker inference cannot establish the validity of the stronger, while resting a weaker inference on the stronger would mean reducing it in which case it would not be distinct. Moreover, these grounds are the logical grounds which depend on the property of symbols and so rest on the same ground as Peirce`s logic in general – so they are metaphysically valid as far as Peirce`s logic is metaphysics.

Now there are exactly three kinds of inference, that is, three kinds of possible symbolization: deduction, hypothesis, and induction. They all belong to the same genus, they are kinds of possible symbolization. But symbolization has three distinct species and thus there is also a specific difference between the inferences, which correspond to the three possible references of a symbol: symbol, form, or thing. Since these are the three references every symbol qua symbol must have, and since inference is symbolization, that is making a symbol out of another, then we can symbolize either (i) a symbol, as in deduction; (ii) a form, as in hypothesis; (iii) a thing, as in induction.

To give examples let us consider these three forms of symbolization.

In the first case of deduction, no new information is added to the term “Galadriel”. In the case of hypothesis, in the second premiss the connotation of “Dr. Mortimer” is “who has forgotten his walking-stick in the detective`s office”, while in the conclusion, its connotation is increased by adding to it the predicate “is absent-minded”. In the case of induction the class of herbivora denotes only “neat and deer”, while in the conclusion it is increased to cover “all cloven-footed animals”.

Peirce then goes on to prove that all three inferential (symbolical) species are valid. Since there are three irreducible kinds of inference where each one has its own leading principles which cannot be reduced to any other. Peirce concludes that each symbolization must be proved by its own leading principles, that is, symbolization of symbol deductively, symbolization of form hypothetically, symbolization of thing inductively. What does however Peirce prove by this procedure? It is not, he says, “that the kinds of inference are valid, since their proof, on the contrary, must rest on the assumption of that validity. Their use is only to show what the condition of that validity is.” (W1: 184–185). What he states is that, paradoxically, there can be no apodictic warrant of the scientific inference. Such a demand would be contradictory – if scientific inference was a priori, it would lose the character which makes it a scientific inference – inferring something not known – and knowable only by scientific reasoning. As we saw, Peirce said, that synthetic inference assumes the validity of some truth which is not given in it. This must be the case for inferring something not yet known, there must be something unknown. Thus, induction and hypothesis cannot be proven apodictically, because, in contradistinction to symbols, they depend upon the symbolization of things and forms, and we can never be sure that we infer correctly in a particular case, what this form or thing is. It is on the other hand guaranteed, that in general, if there is to be any synthetic inference at all, these forms must be valid.

4. Conclusion

We have thus seen some connections which can be found between Peirce`s early metaphysics and his little-later logic, grounded upon the notion of representation. Peirce proposed a logical treatment of metaphysics, which was seen upon its semiotic ground possible, with logical laws have higher authority than the laws of the mind. His definition of “philosophy of primal truths” is mirrored by working-out of the grounds of inference in his logic. Upon the notion of symbolization he shows what are the grounds of inference, and is able to unify synthetic and analytic reasoning in a single theory on the ground of their belong together to the genus symbolization, as well as their specific difference as symbolizing by different leading principles. Such an inquiry is transcendental because if we are to arrive at any knowledge the validity of forms, by which it can be given to us, must be the case without which such knowledge would not be possible. To end up by one last quotation:

[T]o the pupil of Kant as to the pupil of Aristotle the Analytic of Logic is the foundation of Metaphysics. We find oursleves in all our discourse taking certain points for granted which we cannot have observed. The question therefore is what we may take for granted independent of all experience. The answer to this is metaphysics. But it is plain that we can thus take for granted only what is involved in logical forms. Hence the necessity of studying these forms. (W1: 302)

References

KANT, Immanuel. [1787] 2007. Critique of Pure Reason. Translated by Norman Kemp Smith. New York: Palgrave Macmillan.

ESPOSITO, Joseph L. 1980. Evolutionary Metaphysics. Athens, OH: Ohio University Press.

HARTSHORNE, Charles, WEISS, Paul, BURKS, Arthur W. (eds.). 1931–1958. Collected Papers of Charles Sanders Peirce. 8 vols. Cambridge, MA: Harvard University Press.

MURPHEY, Murray G. 1993. The Development of Peirce`s Philosophy. Indianapolis: Hackett.

PEIRCE, Charles Sanders. 1982. The Writings of Charles Sanders Peirce: A Chronological Edition. Vol 1. Bloomington: Indiana University Press.

PEIRCE EDITION PROJECT (eds.). 1992, 1998. Essential Peirce. 2 vols. Bloomington: Indiana University Press.