ABSTRACT: The aim of this paper is to describe the classical conception of rationality, i.e., to indicate the theses traditionally associated with this conception. I do not intend to discuss these theses in detail. Rather, I focus on the question regarding the main elements of the classical conception of rationality. I am interested in the rationality of cognition and of knowledge (epistemic/epistemological rationality), especially in the rationality of science (scientific rationality).

1. Introduction

The aim of this paper is to characterize the classical conception of rationality, i.e. to indicate the theses and conditions traditionally connected with this conception. I do not intend to discuss these theses and conditions in detail. I attempt only to answer the question of what the main elements of the classical conception of rationality are. I limit myself to brief specification and characterization of these elements. The classical conception of rationality that I (re)constructed is an ideal type. Historically known philosophical views and positions are more or less similar to that type. I am interested in the rationality of cognition and knowledge (epistemic/ epistemological rationality), especially in the rationality of science (scientific rationality). I take for granted that the rationality of science is a specific kind of the rationality of cognition and knowledge.

I shall begin with some explanation of the term "classical". Generally, the meaning of this term varies from tradition to tradition and depends also on what is seen as bearers of the property of "being classical". I use the term "classical" in connection with the tradition of ancient philosophy, especially with that of Aristotle and Plato. Among the authors whose conceptions of rationality called here classical I count both the above mentioned ancient philosophers as well as Descartes, Husserl and logical empiricists such as Carnap, Neurath and Reichenbach. A relevant - from the point of view of my problem - element of the classical tradition is its foundational epistemology which is required by the classical conception of rationality. Epistemological foundationalism have extreme and moderate versions. Hence, the conception of rationality may then be strong or weak. In my paper I focus on the radical version of foundationalism. The classical conception of truth and maximalistic approach to justification are these two main ideas of the classical conception of rationality to which all other elements of that conception are subordinated. The primary bearers of rationality are epistemic decisions, for instance the acceptance of a proposition (as true and justified) or its rejection (as false if its falsity is demonstrated). It is rational to accept a proposition if there are good/adequate reasons for it (Aristotle).

2. Theses connected with the classical conception of rationality (constituents of the classical conception of rationality)

In short, the classical conception of rationality is constituted by:

1) demarcationism;

2) the correspondence conception of truth;

3) objectivism/objectivity and realism;

4) theses and requirements concerning (scientific) language

a. the thesis of passive role of language in cognition;

b. the requirement of linguistic precision and exactness of articulation;

5) classical logic

a. the deductive ideal of knowledge;

b. the requirement of consistency and coherence;

c. the requirement of using algorithmic rules;

6) the requirement of justification and criticism;

7) the requirement of accepting basic propositions (premises of reasoning);

8) epistemic values

a. universality;

b. necessity;

c. certainty;

d. absolutism (antirelativism).

1) Demarcationism

Demarcationism (distinctionism) in general is an epistemic position or attitude which consists in looking at the world and human activity in opposite, usually binary categories. Demarcationism which is a component of the classical conception of epistemological (scientific) rationality is an epistemic attitude based on some strong distinctions. The distinctions between rationality and arationality, rationality and non-rationality, between rationality and irrationality (rational and irrational belief/actions) are the most important. For example the decision to accept a proposition is rational if there are adequate reasons (1) for its acceptance and that decision was made on the basis of those reasons. Mathematics and logic provide paradigm examples for rational decisions. A decision is arational if there are no adequate reasons for it and it was taken on the basis of taste, fortune, etc. A decision is non-rational if there are adequate reasons for it but either it was not made or an opposite decision was taken (for example rejection instead of acceptance). A decision is irrational if there are adequate reasons for it but one did not used them and a decision was taken on the basis of inadequate reasons, guess-work being a good example.

Some other important demarcations may be listed: between justified and non-justified propositions; between truth and falsity (true and false propositions); between the context of discovery and the context of justification; between basic and secondary propositions; between universal rules, criteria, principles and solutions of a certain problem, and those particular and culture-determined; between necessary and contingent propositions; between realism and antirealism (idealism); between object and subject of cognition; between language/cognition/knowledge and reality; between objectivity and subjectivity; between absolute and relative knowledge (absolutism and relativism). Some other distinctions may still be added.

Since the concepts used in the above distinctions do not admit degrees, they have thereby classifying character in contrast to vague typological notions. For example, the concept of rationality does not come in degrees: a belief cannot be more or less rational, it is either rational or irrational and tertium non datur. The same concerns the concept of truth.

2) The correspondence theory of truth

Truth in its epistemic sense is a property of propositions or other epistemic/cognitive assertive structures. The classical correspondence theory of truth (where truth belongs to cognition/knowledge) claims that truth is the adequacy of things and the intellect:

To say of what is that it is not, or of what is not that it is, is false; while to say of what is that it is, and of what is not that it is not, is true (Aristotle, Metaphysics 1011b 25).

In other words: "p is a true sentence if and only if p" (A. Tarski). Truth is a property that does not come in degrees, is immutable/invariable and independent of a subject and particular circumstances.

Rationality is an epistemic value secondary to truth. Rational cognitive procedures and rational decisions concerning the acceptance of a proposition should guarantee correct identification of true propositions. Rationality and justification are connected with criteria of truth (e.g. basic propositions are connected with the criterion of evidence), for they are to provide reasons for the acceptance of a given proposition as true. That is the criteria should ultimately distinguish true from false propositions. Justification is what assures that truth of a proposition is correctly determined.

3) Objectivism/objectivity and realism

An objectivist assumes from the very beginning the existence of the epistemic subject-object axis. He believes that ontological difference which exists between the subject and the object is overcame by the act of cognition. A scientistic objectivist maintains that a distance between the subject and the object can be transcended through exact methodical procedures. Knowledge is objective if it is true, i.e. if it corresponds to reality and is not constituted by the cognitive subject. The latter requirement is fulfilled when the subject behaves passively in the act of cognition, i.e. he neutralizes his attitudes, assumptions, wishes, prejudices, feelings, emotions, will, etc., in short - all subjective factors (Husserl, logical empiricists).

Epistemological objectivity is connected with ontological/metaphysical objectivity. In the ontological sense something is objective if it exists independently of consciousness and is such as it is independently of human subjective knowledge. According to the classical conception of rationality the ontological objectivity is the basis for epistemological objectivity and rationality. This conception presupposes then an ontological realism. Cognition and knowledge concern the reality which exists independently of them; the objective external (in relation to the subject) world exists and is cognizable. Its opponents labelled this conception either as the spectator theory of knowledge (J. Dewey), or the copy theory of language (W. V. O. Quine) or finally the metaphor of mind as mirror (R. Rorty).

4) Theses and requirements concerning (scientific) language

a) The requirement of passive role of language in cognition

The linguistic expression of cognitive results is a necessary precondition of any epistemic evaluation of cognition and knowledge. In accordance with the classical conception of rationality the language, being one of the subjective elements of cognition and knowledge, has to play a passive role in cognition. Truth, realism and objectivity of cognition and knowledge are then guaranteed by the passivity of language which permits "mirroring" the reality in knowledge/language. With relation to the reality the language performs descriptive function which is possible to be isolated from other functions (for example, from expressive or valuating functions). The language neither delineates the limits for cognition nor decides about the epistemic view of the world. Neither its conceptual apparatus (vocabulary) nor syntax of the language dictates cognition/knowledge any necessary properties and conditions.

b) The requirement of linguistic precision and exactness of articulation

Linguistic precision is a necessary condition for rationality of cognition and knowledge. In a positive version it is expressed by the requirement that we should use only of precise and exact language, i.e. to use only precise terms and to formulate syntactically unambiguous expressions. In a negative version this requirement says that we should eliminate vague, inexact and ambiguous terms. From the point of view of exactness extensional languages are the ideal of the language (logical empiricist, Lvov-Warsaw school). The requirement of linguistic precision is more rudimentary than the requirement of justification. It is impossible to justify a proposition if it does not have a precise meaning.

5) Classical logic

a) The deductive ideal of knowledge

According to the classical conception of rationality the ideal rational knowledge takes a form of a deductive system. Any deductive system consists of two subsets of propositions: the first includes basic propositions, and the second includes propositions which are deduced from basic propositions. Radical epistemological fundamentalists usually accept the following four theses: 1) knowledge consists of true and justified beliefs; 2) its has a hierarchical structure (system) arranged according to justifying relations; 3) basic statements are certain and do not require justification by other propositions; hence they can be a basis for other propositions; 4) justification of other propositions is mediated by basic propositions.

b) Requirement of consistency and coherence

Rational knowledge conceived as a deductive system should meet the requirement of consistency: two propositions which are inconsistent with each other are not allowed to belong to one system of knowledge. Any acceptance of a system which contains inconsistent propositions is nonrational. (Syntactical) coherence (2) of expression amounts to the property of being correctly built, i.e. according to the rules of a given system.

c) The requirement of using algorithmic rules

Algorithms play a central role in rational decision-making. Rationality of decision to accept a certain conclusion depends not only on the value (truth) of premises of reasoning but also on the adequate rules of inference. Any algorithm provides the set of valid rules of inference, i.e. of valid deductions. Those rules, when applied to a problem, guarantee a solution in a finite number of steps. Algorithms possess the following advantages: 1) they are mechanically applicable; 2) they provide a decision procedure by telling whether a given argument is valid or not; 3) they exclude unreliable factors such as intuition, insight, skill and luck. Once we have applied an algorithm to an argument and found the reasoning valid we have reasons for accepting the conclusion of that reasoning.

Algorithmic rules are used in mathematics when a solution of a problem is determined by providing a proof. The acceptance of that solution depends on the validity of that proof, i.e. on whether the proof conforms to the rules of (classical) logic. In empirical sciences, scientists decide whether a given hypothesis should be accepted by doing proper empirical tests. A decision is rational if it is guided by appropriate rules. The rules of scientific method (for instance induction) determine which tests are relevant, and whether a body of empirical data is sufficient for accepting or rejecting a hypothesis or whether a judgement should be suspended until further investigation.

6) The requirement of justification and criticism

a) Justification

"To justify" (3) a proposition means to provide adequate reasons for its acceptance. Reasons for the acceptance of a given proposition are provided by other propositions we begin with along with rules that establish the connection between those propositions and the proposition in question (K. Ajdukiewicz). In other words, rational acceptance of a proposition (conclusion) is a result of reasoning/argumentation which provides reasons (premises and rules) for accepting that proposition.

Two questions here arise: 1) on what basis do we select propositions from which we are to begin (basic propositions) and 2) on what basis do we select the rules? From a foundational perspective rational justification ultimately requires both premises and rules that can be justified without appealing to other premises and rules. Those premises and rules must be necessary. For if they were not we would face the problem of infinite regress.

b) Criticism

Criticism is a opposite side of the requirement of justification. "To be critical" means not to accept a proposition if there are no adequate reasons for it (Descartes, Hume). However, the requirement "not to accept" is not synonymous with "to reject". Rejection of a proposition requires reasons equally strong as its acceptance.

7) The requirement of accepting basic propositions (premises of reasoning)

Basic propositions are the starting point for reasoning and reasons for acceptance of all other propositions. Reasons for acceptance of basic propositions are not constituted by other propositions for basic propositions are directly justified. They may be justified by means of experience, (sense) perception (logical empiricists), terminological conventions (analytic propositions), (intellectual) intuition (Aquinas) etc. A certain type of evidence is connected with the direct justification of basic propositions.

One way to seek directly justified (self-justifying) propositions is to look for propositions that are self-evident, i.e. propositions that are true and whose truth is clear to anyone who understands them properly (Descartes). (4)

8) Epistemic values

a) Universality

According to the classical conception of rationality both epistemic activities and their results should be universal. Their universality consists in the fact that if a cognitive subject has at his disposal universally applicable algorithmic rules (of reasoning) then everyone who begins with the same premises must arrive at the same conclusion. The rules provide the necessary connections between the starting point and the conclusion. A correct reasoning can lead only to one and the same conclusion. If two cognitive subjects arrive at different results in a particular situation it must be due either to the fact that they do not have the same information, or to the fact that at least one of them is not proceeding in a rational manner. The key idea here is that there is both a definite solution and a definite procedure for arriving at that solution and all who follow that procedure must arrive at the same result. Examples of problems which have one universal solution are provided by mathematics and logic.

In accordance with the classical conception of rationality a belief or decision is rational if it conforms to a set of rules (criteria). If the same rules are applicable in every context, rational scientists need not to debate which criteria should be applied in a particular situation (absence of choice-situation). Universality of both results of epistemic actions and of those actions themselves is then secondary with respect to universality of rules.

b) Necessity

The requirement of the necessary character of basic propositions and conclusions seems to be more fundamental than the requirement of universality. The existence of a necessary relation between the available knowledge (basic propositions) and the rationally acceptable results allows us to understand why every rational cognitive subject who starts with the same premises must arrive at the same conclusion. It is not sufficient that all rational cognitive subjects arrive at the same conclusion since this might occur as a result of massive coincidence, rather than through reasoning (H. Brown). Examples of such necessity are provided again by mathematics and logic. Such issues as what provides foundations for the necessity of relations between propositions in reasoning which in turn guarantees the necessity of conclusions or whether factors which guarantee that necessity are belong to object or subject are here secondary.

c) Certainty

The classical conception of rationality answers also the question of why to be rational. One should be rational for rational procedures provide results which are certain and true (Descartes). It is not sufficient to assert true propositions; one should also possess unquestionable reasons. It is not sufficient to have a right answer to a problem; one must also be able to recognize that a given answer is true if the basis for accepting it is to be provided. Rational procedures give us means for recognizing true propositions. Those procedure not only lead to a conclusion, but they provide us reasons for accepting that conclusion and grounds for believing that it is true. Irrational procedures however do not need to lead to falsehood or error.

Certainty is attributed either to persons as their subjective state or to their beliefs as an inherent property. A belief is accepted in an absolutely certain way when it permits no subjective and objective doubts. The question which here arises is what makes a proposition or a belief absolutely certain and what gives certainty to basic statements. According to the classical conception of rationality this is direct experience/perception which gives that certainty. Experience can be understood in various ways, for instance as sense perception or as intellectual intuition. All other propositions acquire their certainty from basic propositions on the basis of algorithmic rules.

d) Absolutism (antirelativism)

Absolutism of the classical conception of rationality is expressed by the claim that conception and criteria of rationality are invariable and constant (logical empiricists, K. Popper). The principles of rationality are not determined by culture, social circumstances etc. Against relativism it is claimed that there exists a universal difference between rational and irrational beliefs or rational and irrational actions common to all cultures.

Absolutism is connected with the classical conception of truth. A proposition is absolutely true when it is true for everyone, always and everywhere. According to this conception, the absolute criterion of truth admits to judge infallibly whether there is indeed a relation of correspondence between the reality and a given proposition. Absolutism therefore expresses the belief in an indubitable absolute character of some propositions.

3. Final remarks

The theses forming the classical conception of rationality of cognition and knowledge cannot be separated. They intertwine and condition each other. The characteristic of the classical conception of rationality given above shows that rationality is not a simple property of cognition and knowledge. Any conception of rationality is in fact a set of mutually connected theses (conditions) concerning cognition and knowledge.

The classical conception of rationality is normative; it is a project of the ideal cognition and knowledge. The claims held by the adherents of the classical conception of rationality are conditions-norms imposed on cognition, knowledge and truth. The epistemic value realized through fulfilling those conditions is truth classically understood.

Notes

(1) The problem here is what are "adequate" and "inadequate" reasons for accepting or rejecting a proposition.

(2) A different notion of coherence is developed by proponents of the coherence theory of knowledge and justification as well as the coherence theory of truth.

(3) Logical empiricists whose view of rationality can be count among the classical conceptions respect Reichenbach's distinction between the context of justification and the context of discovery and they assume that the questions of rationality arise only in the former. The point of the distinction is that discovery, understood as a process by which cognitive subjects come up with new ideas (propositions), differs essentially from the procedures by which such ideas are tested and evaluated. On this view, everything in the context of discovery is arational: processes and circumstances involved in conceiving a new idea (proposition) are irrelevant to the question of whether we have good reasons for accepting or rejecting that idea. Ideas occur to various people in various situations, and the factors that elicit these ideas can be studied by psychologists, historians, biographers, and so forth. Results of those studies cannot give an answer to the question whether it is rational to accept some idea (proposition).

(4) Self-evident basic propositions are not compatible with objectivism connected with condition of intersubjectivity, because it threatens with solipsism. See discussion within logical empiricism.

Bibliography

Ajdukiewicz K., J_zyk i poznanie, [Language and Knowledge], vol. I and II, Warsaw 1985.

Ajdukiewicz K., Zagadnienie uzasadniania, [Issue of Justification], in: J_zyk i poznanie, [Language and Knowledge], vol. II, Warsaw 1985, 374-383.

Bartley W. W., The Retreat to Commitment, La Salle-London 1984 2.

Bronk A., Filozofia i nauka: problem demarkacji, [Philosophy and Science: Problem of Demarcation], "Roczniki Filozoficzne" 43:1995 nr 1, 181-235.

Bronk A., Zrozumie_ _wiat wspó_czesny, [To Understand the Contemporary World], Lublin 1998.

Bronk A., Majda_ski S., Klasyczno__ filozofii klasycznej, [Classicity of the Classical Philosophy], "Roczniki Filozoficzne" 39-40:1991-1992 nr 1, 367-391.

Brown H. I., Rationality, London, New York: Routledge 1988, 1990.

Haack S., Evidence and Inquiry. Towards Reconstruction in Epistemology, Blackwell (1993) 1995.

Jadczak R., Absolutystyczne stanowisko Kazimierza Twardowskiego w teorii prawdy, [Absolutistic Kazimierz Twardowski's Position in Theory of Truth], in: C. Gorzka, R. Jadczak (red.), Studia z filozofii i logiki. Ksi_ga pami_tkowa w darze Prof. Leonowi Guma_skiemu, [Studies in Philosophy and Logic. Visitors' Book for Professor Leon Guma_ski], Toru_ 1997, 73-85.

Kitcher Ph., The Advancement of Science. Science without Legend, Objectivity without Illusions, New York: Oxford University Press 1993.

Laudan L., Beyond Positivism and Relativism. Theory, Method and Evidence, Westview Press 1996.

Rozum - Racja - Racjonalno__. Materia_y z konferencji naukowej nauk filozoficznych PAN, [Ratio - Reason - Rationality. Materials from Epistemic Conference of Philosophy of PAN], "Studia Filozoficzne" 1983, nr 5-6.

Stiffler E., A Definition of Foundationalism, "Metaphilosophy", 15 (1984), nr 1, 16-25.

Triplett T., Recent Works on Foundationalism, "American Philosophical Quarterly" 27 (1990), 93-116

Walker R. C. S., Theories of Truth, in: B. Hale, C. Wright (ed.), A Companion to the Philosophy of Language, Oxford & Cambridge: Blackwell Publishers 1997, 309-330.

Wole_ski J., Philosophical Lvov-Warsaw School, [Filozoficzna szko_a lwowsko-warszawska], Warszawa 1985.