Epistemology: Basic Terms, Concepts & Views

Epistemology: Basic Terms, Concepts & Views Prof. Dr. Kenneth R. WESTPHAL Department of Philosophy (Felsefe) Boðaziçi Üniversitesi (Ýstanbul) Rev. 15.09.2016

NOTE: Although these notes most directly concern knowledge of objects, most of the central issues concern rational justification, empiricist concerns about rational justification have been so central to Modern and Contemporary philosophy, that many of the issues discussed here about rational justification also pertain to moral philosophy. (It is worthwhile to consider which of these issues do, and which do not, pertain to moral philosophy, and how they do, or why they do not.)

1

Knowledge & Justified True Belief

3

2

Criteria vs Nature of truth

3

3

Knowledge a priori & a posteriori

4

4

Synthetic vs analytic statements

4

5

Realism, Idealism, Relativism & Scepticism

4

6

Scientia vs Historia

5

7

Fallibilism vs Infallibilism

5

8

Justification & Scepticism

6

8.1

The Regress Argument

6

8.2

Epistemic Principles: The regress of levels

7

8.3

Foundationalism

7

8.4

Self-Justifying Basic Claims: Logical & Substantive

8

8.5

Coherentism

8

9

Empiricism

9

9.1

Concept Empiricism

9

9.2

Verification Empiricism

9

9.3

Meaning Empiricism

10

10 Rationalism

10

10.1

Concept Rationalism

10

10.2

Rationalism regarding Justification

10

10.3

Rationalist Foundations: A Priori Principles

10

1

10.4

Empiricist Foundations: Experiential Claims

11 Pragmatism

11 11

11.1

Pragmatism about Meaning

11

11.2

Pragmatism about Justification

12

12 Causal Reliability Theory

12

13 Information Theory

12

14 Historicism & Relativism

13

14.1

The Anti-foundationalist’s Trilemma

15 Internalism & Externalism

13 13

15.1

Mental Content Internalism

14

15.2

Semantic Internalism

14

15.3

Justificatory Internalism

14

16 Individualism & Collectivism

14

16.1

Atomistic Individualism

14

16.2

Monolithic Collectivism

14

16.3

Moderate Collectivism

15

17 Reason & Tradition

15

18 Hermeneutics & Hermeneutic Circularity

15

18.1

Textual Hermeneutics: Schleiermacher

16

18.2

Ontological Hermeneutics: Heidegger

16

2

1

KNOWLEDGE & JUSTIFIED TRUE BELIEF.

Here’s a simple SCHEMA FOR UNDERSTANDING KNOWLEDGE: Knowledge = Justified T rue B elief (‘JTB’) The rationale for including each of these elements in an account of knowledge is fairly straightforward. BELIEF: Knowledge concerns, not simply the facts, or how things are in the world. It concerns our knowledge of or ideas about those things. We have lots of beliefs, and we are familiar with many other beliefs that other people have. Some beliefs are fantasies, others are vague, and some seem to be accurate. To have knowledge, one must (it seems) at least believe something to be the case. Instead of beliefs, we might consider judgments or claims. The main point is that knowledge is something of which people have some command; in contrast to information catalogued in documents, people have knowledge. TRUTH: Believing something to be the case does not suffice for knowledge, however. There is a distinction between wishful thinking and knowledge, or between either of these and opinion. One part of that distinction lies in the idea that, to count as knowledge, a belief must be true – and not simply held with conviction. JUSTIFICATION: While important, truth also does not suffice for knowledge. There’s a distinction between knowledge and lucky guessing. A lucky guess might be both believed and be true. Is that all there is to knowledge? Is this how people go about trying to get knowledge – by trying to make lucky guesses? Well, for many matters people try to collect evidence. Evidence is important because it provides justification; it helps show that some beliefs about a particular issue are more likely to be true than other beliefs about it. Most of the controversy in epistemology has concerned what kinds of justification there may be, and how much of any of it suffices for knowledge. These considerations lead to the ‘JTB’ schema for knowledge. The justification must pertain to truth, knowledge and evidence; i.e., it must be cognitive justification, in contrast, say, to mere utility. Often philosophers speak or write of ‘epistemic’ justification, where specifically cognitive justification is at issue. I strongly prefer calling it cognitive justification, and reserve ‘epistemic’ justification for the justification of theories of knowledge. In this, I follow William Alston’s important cautions against confusing levels of analysis within epistemology. 2

CRITERIA VS NATURE OF TRUTH. TRUTH:

A statement is true if it accurately represents something, by ascribing characteristics to something which in fact has those characteristics.

The above definition purports to characterize the nature of truth; it doesn’t tell us how to identify which statements are true and which aren’t. Statements are true or false; beliefs or knowledge are justified or unjustified. Truth requires both reference and ascription. Analogously for beliefs (contents of beliefs formulated as statements). CRITERION of truth:

Something which helps distinguish true from false (or more from less accurate) statements. 3

There’s been much discussion of criteria of truth, but the core issues about criteria of truth turn on issues about sufficient (and sufficiently accurate) justification, and criteria of justification. CRITERION of justification:

Something which helps distinguish sufficiently justified from insufficiently justified, or better from less well justified statements.

Issues about justification are discussed throughout the remainder; many of these issues also pertain to moral philosophy. 3

KNOWLEDGE A PRIORI & A POSTERIORI.

The basic distinction between knowledge a priori and a posteriori is this: A posteriori Knowledge:

Any proposition or claim which is known only on the basis of sensory or experiential evidence is known a posteriori; it is known ‘posterior’ to or after sensory experience.

A priori Knowledge:

Any proposition or claim which can be known without relying on sensory or experiential evidence is known a priori, or ‘prior to experience’.

There’s an important wrinkle to pay attention to here: It may be the case that some amount or kind of experience is necessary simply to understand the terms of a proposition, including one supposedly known a priori. The point about ‘a priori’ knowledge is that one needs no further experience or evidence, beyond whatever is necessary for understanding the proposition, to know whether the proposition is true or false. 4

SYNTHETIC VS ANALYTIC STATEMENTS.

The basic distinction between synthetic and analytic statements or propositions is this: Analytic statements are either self-contradictory (and so are necessarily false), or else their negations are self-contradictory (in which case, those statements are necessarily true). Synthetic statements are logically consistent, and their denials (negations) are also logically consistent. Combining these two distinctions gives us four possibilities, presented in the following chart: K n o w l e d g e

Statements: Analytic

Synthetic

/ a priori ___________/__________ a posteriori

/

/

Which of these combinations represent plausible or possible classifications of human knowledge, and which are empty?

4

5

REALISM, IDEALISM, RELATIVISM & SCEPTICISM. REALISM:

The objects of knowledge (and their characteristics) are what they are regardless of what we think or say about them.

This is often called ‘common sense realism’. Frequently, this idea is put in terms of the objects of knowledge being ‘independent’ of our knowledge of them. Notice, however, that this way of speaking tends to cause confusion: if objects of knowledge are literally ‘independent’ of our knowledge, then isn’t scepticism about such knowledge justified? (Or: what sorts of ‘independence’ matter to these issues, and which do not?) IDEALISM:

The objects of knowledge (or their characteristics) depend upon the character of the mind which knows the(m Th. is is one standard sense of the term ‘idealism’.)

RELATIVISM: What there is, or what is true, is relative to (depends on) some set of minds, either individually or collectively. SCEPTICISM: or: 6

No one knows anything (about __________). So far, despite their efforts, analysis and evidence, no one appears to know anything (about _________).

SCIENTIA VS HISTORIA. Scientia:

Only strict deduction suffices for rational justification. This alone provides for ‘genuinely scientific’ knowledge, in contrast to (perhaps very well grounded) conjecture or opinion.

Historia:

Inquiry and all results of inquiry, where ‘inquiry’ contrasts to scientia and involves investigating empirical phenomena and assessing empirical evidence (of whatever kind, whether natural, social or historical).

This distinction traces back at least to Aristotle, but scientia is equated to strict deductivism only in the 13th Century (C.E.); in this strong form, it is expressly invoked, e.g., by Descartes, Locke and Hume. (Kant knew the distinction, though his use of it is complex.) Scientia is central to Cartesianism in all its forms, including much of empiricism and of analytic epistemology and philosophy of mind. Sometimes this distinction is marked by the contrasting terms ‘rational’ and ‘historical’ knowledge. The distinction between scientia and historia is directly linked to that between a priori and a posteriori knowledge, as the distinction between what can be known without, and what can only be known (if at all) as the result of (empirical) inquiry. There is a crucial philosophical and historical point lurking here: Aristotle’s model of scientific knowledge (scientia) is modelled on Euclidian geometry, yet it is expressly a flexible model, to be adapted to suit the relevant domain of the science in question. This flexible model of scientia becomes strict, infallibilist deductivism in March 1277, when Etienne Tempier, acting on authority of the Roman Pope as Bishop of Paris condemned 220 neo-Aristotelean theses as heretical.1 His condemnation declares and in many comments also implies that knowledge is only of demon1

See David Piché (1999), La Condamnation parisienne de 1277 (Paris, Vrin); Stephen Boulter (2011), ‘The Medieval Origins of Conceivability Arguments’, Metaphilosophy 42.5:617–641.

5

strative truths, the negations of which are self-contradictory. This view merely awaits Descartes generalisation of this requirement into the possibility of global perceptual scepticism. 7

FALLIBILISM VS INFALLIBILISM. This distinction pertains to theories of justification.

8

INFALLIBILISM:

Justification of a claim (belief, statement, judgment) sufficient for that claim (etc.) to count as knowledge entails the truth of that claim.

FALLIBILISM:

Justification of a claim (etc.) sufficient for that claim (etc.) to count as knowledge does not entail the truth of that claim.

JUSTIFICATION & SCEPTICISM.

Because philosophers are concerned with settling substantive disputes in various areas (philosophy of science, metaphysics, philosophy of religion, ethics), demonstrating that one knows something (some substantive claim) has been the central epistemological concern. Demonstration is one kind of justification, and scepticism has been central to epistemology because it raises crucial issues about justification. The concern with demonstrating various claims to be true lies not only with knowing substantive claims (about ethics, God, mater, etc.), but with knowing that one knows those claims or with showing that one knows those claims. Showing that or knowing that one knows something is a tricky affair, for those who disagree and especially those who are sceptics generally can and will press for a justification for any claim you make, including the last claim you made to justify your original claim. How can one answer the question, ‘Well, how do you know that?’ What could bring this series of questions to an end? Well, the series might end in a false claim. Such a terminus is no help in showing that one knows something. The series might end with an assumption or a mere assertion, but this, too, is no help. The series might instead form a loop, where one repeats an earlier claim at a later point in one’s regress of justification. This, too is no help. This problem is generated by what is called the ‘regress argument’, an argument propounded by a sceptic which asks for justification for each claim you might make, and for each claim which you make in justifying your last claim, etc. This argument was classically formulated by Sextus Empiricus: 8.1 The Regress Argument: ‘The [second] Mode based upon regress ad infinitum is that whereby we assert that the thing adduced as a proof of the matter proposed needs a further proof, and this again another, and so on ad infinitum, so that the consequence is suspension, as we possess no starting-point for our argument ... We have the [fourth] Mode based on hypothesis when the Dogmatists, being forced to recede ad infinitum, take as their starting-point something which they do not establish by argument but claim to assume as granted simply and without demonstration. The [fifth] Mode of circular reasoning is the form used when the proof itself which ought to establish the matter of inquiry requires confirmation derived from that matter; in this case, being unable to assume either in order to establish the other, we suspend judgment about both’. (Sextus Empiricus, Outlines of Pyrrhonism 1:166–9.)

Now what could stop this regress? Notice that only a very special kind of claim or belief could stop the regress. The regress is generated because each claim or belief must not only be justified, but must be shown or known to be justified. Usually showing or knowing that a claim or belief is justified requires appealing to independent grounds, grounds other than that belief or 6

claim. However, if one appeals to independent grounds, then the sceptical question recurs: How are those grounds known? The regress could only be stopped by a claim or belief which shows its own justification, and so cannot possibly be mistaken. Such claims or beliefs may be called ‘self-justifying’. The Modern epistemological concern with answering the sceptical regress argument leads most modern and contemporary philosophers to adopt a ‘foundationalist’ epistemology. This sense of ‘foundationalism’ is that all knowledge of the world (empirical knowledge) rests on ‘foundations’, basic, self-justified claims or beliefs, from which all other empirical knowledge follows (either deductively or inductively). This kind of view is very strong, and so especially difficult to develop plausibly. Claims that are both self-justifying and substantive – substantive enough to support ordinary or scientific claims to knowledge – are very rare. Consequently, many weaker forms of foundationalism have been developed which require weaker conditions on basic claims or foundations. I’ll come back to this shortly. 8.2 Epistemic Principles: The regress of levels. The problem about what kinds of knowledge there are, or about what roles are played by various kinds of claims or evidence in our knowledge, introduces another very important aspect of the Regress Problem. The Regress Problem focuses on our search for evidence for our beliefs. However, our search for evidence requires that there can be such a thing as evidence; that is, that some worldly state of affairs or some psychological state could have a relation to some belief such that the state of affairs or the psychological state is evidence for that belief. In general, this issue concerns epistemic principles: What are the principles of justification whereby something counts as justifying evidence for a belief or a claim to knowledge? Moreover, how are those principles of justification themselves justified? Here is another, closely related, regress. This time the regress is a regress of levels of justification: Justifying some belief by appeal to some bit of (alleged) evidence requires appealing to some epistemic principle according to which that bit of whatever does count as evidence for that belief. Yet to show that the belief in question is justified by that evidence, by appeal to that principle, requires also showing that the principle is justified, and showing that principle is justified requires appeal to some further, higher level principle, but that appeal can only be admitted if that further, higher level principle is justified, etc. Let’s make this problem more concrete by discussing one obviously relevant (if also generic and bland) epistemic principle: The Reliability of Perception:

Sense experience is (usually) a reliable source of perceptual beliefs.

How can such epistemic principles be known? If they are to be known, it must be on the basis of adequate reasons. Principles such as these are not self-evident; their denials are not self-contradictions, and they certainly could be believed without being true. They are not ‘self-warranted’ in the way that, e.g, claims about one’s current conscious states are self-warranted. And they are not themselves justified in any ‘direct’ way as are simple perceptual beliefs. Thus there is no alternative to having adequate reasons for justifying such epistemic principles. The problem, then, is to understand what sorts of reasons could be given to justify epistemic principles, and how those reasons can be adequate to justify them. The problem with justifying an epistemic principle about the reliability of perception is that we apparently have no way of learning about perception or its reliability other than by relying on actual sense perception. This 7

threatens to issue in circularity, which threatens to undermine any possibility of justifying principles about the reliability of perception. 8.3 Foundationalism: 1

There are two kinds of knowledge, basic and derived.

2

Any and all bits of basic knowledge can be had independently of any other bit of knowledge. The justification of any bit of basic knowledge is thus independent of other knowledge one may have (or think one has).

3

All other knowledge – derived knowledge – is justified by deriving it from basic bits of knowledge.

Central to foundationalism is the asymmetrical relation between basic knowledge and derived knowledge indicated in the third tenet (3). Justification is provided by basic knowledge for derived knowledge, not vice versa. This foundationalist schema can be filled out in many ways. There are various candidates for each kind of knowledge (basic & derived), and there are various candidates for the kind of justification each kind of knowledge may have. We’ll discuss some of these options during the semester. 8.4 Self-Justifying Basic Claims: Logical & Substantive. Self-Justifying Basic Claim:

A basic claim is self-justifying if making the claim (or having the belief) in question suffices to understand the claim or belief and to know that the claim or belief is true.

What sorts of claims or beliefs could be ‘self-justifying’ or ‘self-evident’ in this way? Well, perhaps simple mathematical truths, such as ‘2 + 2 = 4’. Perhaps such claims are self-evident (although Descartes argues they are not). What about simple logical truths, such as ~ (A & ~ A)? Such claims might also be self-evident. However, these claims don’t help solve the problem. The problem was defending knowledge of the world, and mathematical and logical truths can be true without anything else existing at all. What we need are self-evident truths about existing things. The Modern period offers two different kinds of self-evident truths about existing things. The main rival to foundationalism is coherentism. Basically, coherentism denies that there is any such ‘basic’ knowledge as foundationalists maintain. 8.5 Coherentism: 1

There are no self-evident foundations of knowledge.

2

There is no fundamental distinction between basic and derived knowledge.

3

Any piece of knowledge is justified by recourse to other pieces of knowledge.

Coherentism emphasizes the interdependence of our beliefs, of our use of evidence, and of the concepts we use in formulating our judgments or beliefs. This is a prominent theme among pragmatists, and it comes to be a prominent theme in later analytic philosophy (Quine).

8

One common (20th-Century) argument for coherentism: (1) There are no self-evident foundations. (2) Sensory states aren’t propositions. (3) Logical implications hold only among propositions. (4) Justification requires logical implication. ˆ (5) Justification can only be a matter of relations among propositions. ˆ (6) The only relation that could do the job of justifying a proposition is systematic mutual implication among the most comprehensive system of propositions possible.

A Problem: Can there be equally comprehensive, internally consistent, but mutually inconsistent systems of propositions, so that all of them would be equally justified by any coherence theory of justification, and hence each of them is shown to be true, even though they are mutually inconsistent and so cannot all be true (of the same objects at the same time in the same regard(s), as Aristotle reminds us)? 9

EMPIRICISM.

David HUME (1711–1776) is the paradigmatic empiricist; analytic epistemology began when modern symbolic logic was developed and then used to try to carry through Hume’s program by replacing Hume’s weak psychology with Russell & Whitehead’s Principia Mathematica. Hume’s official view is an early version of Ayer’s Language, Truth, and Logic. The most sophisticated and insightful 20th-century empiricist was Rudolf Carnap. Hume’s three most important theses (for our purposes) are ‘concept empiricism’, ‘verification empiricism’, and ‘meaning empiricism’. 9.1 Concept Empiricism: 1

Every complex term in our language (i.e., every complex concept) can be defined in terms of the simple terms of our language, where each simple term designates some one simple perceptual quality. (This is an account of semantics, but not syntax; see below.)

2

There is no Cartesian cogito, whereby we obtain knowledge of ourselves as substances, or through which we obtain the concepts of substance or of self.

3

Somehow, we get logical concepts (& syntax).

The first tenet (1) is complex because it embeds a complex view about meaning and about philosophy of mind. The most important points are these. First, concept empiricists strongly tend to be nominalists; according to nominalism, concepts just are terms which are used in certain ways. Regarding ‘simple’ terms, concept empiricists typically appeal to sensory atomism. Locke states this view very lucidly, saying that sensory qualities (‘sensory ideas’, he calls them) produced as we sense some object in our environs enter our senses ‘simple and unmixed’, that is to say, each ‘simple’ (pure, uncomplex) sensory quality is distinct from all other sensory qualities, both qualitatively and quantitatively; i.e., they are numerically distinct, and thus are sensory counterparts to atoms.2 Their second basic idea is that a ‘complex’ term designates some com 2

Locke states: ‘The better to understand the nature, manner, and extent of our knowledge, one thing is carefully to be observed concerning the ideas we have; and that is, that some of them are simple, and some complex. ‘Though the qualities that affect our senses are, in the things themselves, so united and blended, that

9

plex of simple sensory qualities. Third, the term ‘semantics’ is vague because it is used in various senses. Amongst philosophers, when contrasted with ‘syntax’, ‘semantics’ concerns reference of terms to objects (or events, or qualities), and concerns ‘meaning’ insofar as a term ‘means’ what it refers to or designates. ‘Syntax’ concerns formation rules for combinations of terms into proper sentences. The most familiar examples of syntactic formation rules are grammatical. Concept empiricists typically specify further, more restricted rules to insure that any genuinely meaningful sentence accord with concept empiricism, as stated just above. 9.2 Verification Empiricism: For any proposition, one can tell whether it is true or false, or one can justify it, in only one of two ways: 1

it’s true by definition.

(Hume’s ‘relations of ideas’.)

2

it is justified (at least in principle) by experiential, sensory evidence. (Hume’s ‘matters of fact’.)

9.3 Meaning Empiricism: Any proposition is meaningful only if it meets either of two conditions: 1

it is necessarily true (it’s denial is self-contradictory)

2

it can be verified by some set of sensory experiences.

(Hume’s ‘commit it to the flames’.) Hume’s aim, like the aim of later postivists, was to clear away all metaphysics, rationalist ethics, and religion, and leave only common sense, science, and mathematics. According to Hume, only two boxes have any content: the only propositions which can be known a priori are analytic; all synthetic propositions can only be known (if at all) a posteriori. In this regard, Hume is a paradigmatic empiricist. 10 RATIONALISM. Modern epistemology began with Descartes’ Meditations, which raised epistemology to the status of first philosophy, thus displacing metaphysics or, as Aristotle put it, ‘the study of being qua being’. Descartes was a rationalist and a foundationalist. 10.1

Concept Rationalism: Some meaningful terms in our language (i.e., some concepts) cannot be defined in terms of the simple terms of our language, where these simple terms refer to, name, and ‘mean’ simple perceptual qualities.

10.2

Rationalism regarding Justification: Some synthetic propositions can be known a priori, without relying on sensory experience.

An important detail here needs to be clarified. It may be that a certain amount of experience is necessary to understand the terms used in a synthetic proposition, including a synthetic proposithere is no separation, no distance between them; yet it is plain, the ideas they produce in the mind enter by the senses simple and unmixed. For though the sight and touch often take in from the same object, at the same time, different ideas; as a man sees at once motion and colour; the hand feels softness and warmth in the same piece of wax; yet the simple ideas, thus united in the same subject, are as perfectly distinct as those that come in by different senses: The coldness and hardness which a man feels in a piece of ice being as distinct ideas in the mind, as the smell and whiteness of a lily; or as the taste of sugar, and smell of a rose. And there is nothing can be plainer to a man, than the clear and distinct perception he has of those simple ideas; which, being each in itself uncompounded, contains in it nothing but one uniform appearance, or conception in the mind, and is not distinguishable into different ideas. (Essay Concerning Human Understanding, Bk. 2 Ch. 2 §1)

10

tion which can be known a priori. Knowing what a term means, and knowing what a synthetic proposition means on the basis of its terms, does not suffice for knowing whether the proposition is true. Rationalists hold that, once we have enough experience to understand what some synthetic propositions mean, we can then prove a priori that they are true. That is, we don’t need any further sensory evidence to verify the truth of that proposition. 10.3

Rationalist Foundations: A Priori Principles.

Rationalists tend to take recourse to a priori rational principles. Descartes attempted to rebuild the whole of empirical knowledge based on the claim, ‘I think, [hence] I am’. This is a claim about an extant object – the speaker or thinker who utters the statement – and it cannot possibly be false any time it is uttered or thought. This is a very interesting claim; the problem is getting from this claim (the so-called ‘cogito’) to the knowledge of anything else. What about the principle, ‘every event has a cause’? Many thinkers in the Modern period thought that this principle is a necessary truth, like mathematical or logical principles. Its denial was thought to be self-contradiction. Descartes believed this, and he used this principle to demonstrate the existence of God, and from there, to justify empirical knowledge. Hume showed conclusively that this principle, ‘every event has a cause’, is not a necessary truth. Consider the following two statements: ‘Every effect has a cause’.

‘Every event has a cause’.

The first of these is a necessary truth; an effect just is whatever results from the activity of a cause. The second is not a necessary truth; it’s denial is not a self-contradiction. As Hume showed, it is logically possible (even if, perhaps, it is physically impossible – at least at the macrolevel) for something to just pop into or out of existence; no cause, no explanation, no knowledge of how it happened – but also no logical contradiction. Rationalist principles alone are not sufficient to justify empirical knowledge. 10.4

Empiricist Foundations: Experiential Claims.

What about perceptual claims or beliefs? Are there any self-evident or self-justifying perceptual beliefs? Not of any ordinary sort. What about our perceptual beliefs that there is a black board here? It seems plain enough that we see it, it’s black, it’s the right size, shape, and location. I can testify that it’s hard, and I’m sure I could get it to make those horrid screeching sounds with my finger nails. Looks like a good perceptual belief. But is it self-justified? Hardly. What if suddenly it started changing colours, or if it started emitting the Star Spangled Banner? That would show that it wasn’t a black board after all, and that our belief that it is was false. Perceptual beliefs about ordinary physical objects aren’t self-justifying. If they were, there wouldn’t have been any sceptical problems to begin with. This has led philosophers in both the Modern and the Contemporary period to suppose that there are some special sorts of sensory states, such as ‘it seems to me that (or: it appears to me that) there’s a blackboard here’. Such a claim is self-evident; if something seems to you to be the case, then it certainly does seem to you to be the case. The problem with these special states (appearances, seemings, sense data) is that they don’t entail or otherwise justify any of the ordinary claims we think we know, and certainly not any of the less ordinary claims that have been subject to dispute among philosophers or among other reasonable thinkers. 11

11 PRAGMATISM. Pragmatism is an alternative to foundationalism and to coherentism. Wilfrid Sellars succinctly characterises the distinctive point of pragmatism: Above all, the [foundationalist] picture is misleading because of its static character. One seems forced to choose between the picture of an elephant which rests on a tortoise (What supports the tortoise?) and the picture of a great Hegelian serpent of knowledge with its tail in its mouth (Where does it begin?). Neither will do. For empirical knowledge, like its sophisticated extension, science, is rational, not because it has a foundation but because it is a self-correcting enterprise which can put any claim in jeopardy, though not all at once. (Wilfrid Sellars, ‘Empiricism & the Philosophy of Mind’, in: Science, Perception and Reality [London: Routledge & Kegan Paul, 1963], 170)

11.1

11.2

Pragmatism about Meaning: 1

A term or idea is meaningful only insofar as it has a use and only when it is used.

2

The meaning of a term or idea is the difference it makes to those activities which involve its use.3

Pragmatism about Justification: so long as

A statement is justified – provisionally – insofar and

1. It is more adequate to its tasks than any alternative statement; 2. It is adequate to its designated range of use or phenomena; & 3. It remains adequate to its designated range of use or phenomena as its use is renewed upon new, relevant occasions, which may included changed circumstances or context. Pragmatists are thus fallibilists about justification. 12 CAUSAL RELIABILITY THEORY. Reliabilism: A true belief counts as knowledge if it is reliably produced or causally sustained by the fact the belief is about. The failure of foundationalism has led recent philosophers to try to understand knowledge in terms of reliable generation of beliefs through physical and neuropsychological processes, understood either as causal chains or as information channels. The idea is that, e.g., perceptual beliefs are ‘reliable’ if they are produced by processes which generate any such belief only when 3

Peirce goes so far as to insist that, no matter what other sorts of differences of attitudes or feelings we may associate with our beliefs, the only genuine differences between beliefs are differences in the actions they guide: ‘The essence of belief is the establishment of a habit, and different beliefs are distinguished by the different modes of action to which they give rise. If beliefs do not differ in this respect, if they appease the same doubt by producing the same rule of action, then no mere differences in the manner of consciousness of them can make them different beliefs ...’ (‘How to Make our Ideas Clear’ [in M. White, The Age of Analysis], 144); ‘Thus, we come down to what is tangible and practical, as the root of every real distinction of thought, no matter how subtle it may be; and there is no distinction of meaning so fine as to consist in anything but a possible difference of practice’ (145). Ideas and their content are, Peirce maintains, strictly tied to the effects of things experienced during the activities in which we engage those things: ‘Our idea of anything is our idea of its sensible effects ...’ (146); ‘Consider what effects, which might conceivably have practical bearings, we conceive the object of our conception to have. Then, our conception of these effects is the whole of our conception of the object’ (146).

12

that belief is true; e.g., ‘I see my computer screen in front of me now whilst I type this silly example’. There have been many problems analysing the precise conditions such causal processes or information channels must meet to be suitably reliable. These sorts of reliable causal or informational origins provide a plausible way in which to understand how we could be justified in various perceptual beliefs. However, they have little to say about how anyone could show to someone else that he or she has a justified perceptual belief; similarly, they don’t suggest how we could know that we know something. Finally, such theories often have trouble accounting for false beliefs. 13 INFORMATION THEORY. A sophisticated successor to causal reliability theory was developed by Fred Dretske, by developing a semantic theory of information.4 It is too subtle to summarise here; I mention it to forestall the common assimilation of Dretske’s information theoretic epistemology to garden variety causal reliability theory. Dretske (1981) demonstrates that causal relations are neither necessary nor sufficient for information relations, and that information relations are necessary (if not sufficient) for any natural occurrence – including responses of our sensory channels – to convey the kind(s) of informational content required for any specifically semantic content (meaning, intension), where semantic content is necessary for beliefs or statements and so for specifically cognitive states or knowledge. 14 HISTORICISM & RELATIVISM. It has been widely supposed that upholding realism in epistemology requires rejecting a social account of knowledge. If what one ‘knows’ depends on one’s society and its particular historical circumstances, then that supposed knowledge would seem to be a function of one’s society (say, a function of the language and beliefs of that society) rather than being a function of the things supposedly known. A socially grounded theory of knowledge thus appears to lead to ‘historicism’. Historicism is a version of ‘relativism’ according to which the truth is relative to a particular historical society or age. Thus historicism is incompatible with realism. Consequently, upholding realism has widely been regarded as requiring an individualist account of human knowledge: in principle, anything which we can know, we can know independently of anyone else, including the entirety of our society. (Witness Descartes’s solo ‘meditations’.) For a variety of reasons which we’ll examine in due course, foundationalism has fallen out of favour with most epistemologists. If foundationalism is untenable, then we’re left confronting a trilemma: 14.1

The Anti-foundationalist’s Trilemma: Can the following 3 questions all be answered conjointly (together) and affirmatively? 1) Is there a way the world is regardless of how we think about it? 2) If so, can we know the way the world is? 3) Is knowledge a socio-historical phenomenon?

4

Dretske, Frederick I., 1981. Information and the Flow of Knowledge. Cambridge, Mass., MIT/Bradford Books. ———, 1995. Naturalising the Mind. Cambridge, Mass., MIT/Bradford Books. ———, 2000. Perception, Knowledge, and Belief: Selected Essays. New York, Cambridge University Press.

Various philosophers in various periods have answered these questions differently, but rarely has it been thought that all three questions could be answered affirmatively. Scepticism holds that there may be a way the world is, but we can’t know it. ‘Subjectivism’ holds that we can know the way the world is, but its structure or characteristics depend upon our cognitive or linguistic activity. Realism in epistemology holds that we can know the way the world is, even though it is not dependent upon our cognitive or linguistic activity. Almost all realists have held, however, that realism in epistemology requires rejecting the view that human knowledge is in any way historically and socially based. Relativists and Historicists have agreed with this disjunction, but have converted it modus tollens and argued that precisely because knowledge is a social and historical phenomenon, we must accept subjectivism and reject realism, we must reject the idea that what we know is what is it independently of our thoughts or language. There is thus a widely supposed dichotomy between realism & social or historical bases of cognition. Pragmatists, too, disagree among themselves about subjectivism and realism, and so do their expositors. Richard Rorty rejects realism altogether, and does so in the name of John Dewey. I agree with Ralph Sleeper that Rorty is wrong philosophically, and wrong about Dewey. That dispute aside, Peirce, the late Dewey, Sellars, and most of our authors definitely hold that this is a false dichotomy: social & historical dimensions of knowledge are compatible with realism. That will be a recurrent theme in our module. 15 INTERNALISM & EXTERNALISM. With increasing critical, and increasingly critical discussion of ‘Cartesianism’ in contemporary philosophy initiated by Burge (1979), much discussion has turned on distinctions between ‘internalism’ and ‘externalism’ regarding semantics, mental content or cognitive justification.5 The basic distinctions in each case are the following, where the Cartesian option is internalist. 15.1

MENTAL CONTENT INTERNALISM: Any and all contents of awareness, or any and all ‘mental’ content, can be fully specified without reference to anything non-mental, in particular, without reference to any features of one’s physical environment, or to features of one’s merely somatic (bodily) states. MENTAL CONTENT EXTERNALISM: At least some ‘mental’ contents can be fully specified only by reference to non-mental phenomena, especially, by reference to objects or events in one’s physical environment, or perhaps to features of one’s merely somatic (bodily) states.

15.2

SEMANTIC INTERNALISM: Any and all conceptual content or linguistic meaning (descriptive content, intension) can be fully specified without reference to anything nonmental, in particular, without reference to any features of one’s physical environment, or to features of one’s merely somatic (bodily) states. SEMANTIC EXTERNALISM: At least some conceptual content or linguistic meaning (descriptive content, intension) can be fully specified only by reference to non-mental, non-linguistic or non-conceptual phenomena, especially, by reference to objects or events in someone’s physical environment, or perhaps to features of someone’s merely somatic (bodily) states.

5

See Tyler Burge (1979), ‘Individualism and the Mental’, Midwest Studies in Philosophy IV: Studies in Metaphysics

(1979):73–121; Origins of Objectivity, Oxford University Press, 2010.

14

15.3

JUSTIFICATORY INTERNALISM: Any and all factors bearing upon the cognitive justification of someone’s (S’s) beliefs, claims or judgments must be ‘internal’ to S’s ‘conscious perspective’, in the sense that any and all factors bearing upon the cognitive justification of S’s beliefs, claims or judgments must be such that S is aware of them, or can easily become aware of them by simple (uncomplicated, unsophisticated) reflection. JUSTIFICATORY EXTERNALISM: NOT all factors bearing upon the cognitive justification of S’s beliefs, claims or judgments must be such that S either is, or can easily become aware of them by simple reflection.

16 INDIVIDUALISM & COLLECTIVISM. It is often supposed that there are only two possible views of the relation between individuals and their societies: 16.1

Atomistic Individualism: Individuals are ‘prior to’ or more basic than society, in the sense of being more basic than society, so that societies are just collections or aggregations of individuals.

16.2

Monolithic Collectivism: Society is ‘prior to’ or more basic than its individual members, in the sense that particular individuals are so deeply molded by their societies that they are nothing but members or components of it.

However, there is a third alternative: 16.3

Moderate Collectivism: 1

Individuals are fundamentally social practitioners. Everything one does, says, or thinks is formed in the context of social practices that provide material and conceptual resources, objects of desire, skills, procedures, and the like.

2

What individuals do depends on their own response to their social context.

3

There are no individuals, no social practitioners, without social practices, and vice versa, there are no social practices without social practitioners; without individuals who learn, participate in, perpetuate, and who modify those social practices as needed to meet their changing needs, aims, and circumstances.

If moderate collectivism is true, then the standard debate between atomistic individualism and monolithic collectivism is spurious. 17 REASON & TRADITION. A wide-spread assumption of the Enlightenment, especially prominent in Descartes and still influential today, is the notion that one key aim and purpose of reason is to assess critically the content and status of traditionally acquired beliefs and knowledge. To perform this critical role, it is often presumed, reason must be independent of tradition. Hence reason must be an inherent endowment of individuals, and not a social phenomenon. Consider whether or how the three views on individualism and collectivism just mentioned bear on this assumption, and how this assumption bears upon them. 15

18 HERMENEUTICS & HERMENEUTIC CIRCULARITY. 18.1

Textual Hermeneutics: Schleiermacher – see: Michael Inwood (2016), ‘Hermeneutics’, §2. In: E. Craig, ed., Routledge Encyclopedia of Philosophy. London, Routledge; DOI: 10.4324/9780415249126-P023-2.

18.2

Ontological Hermeneutics: Heidegger – see: (ibid.), §4. * * *

16