green numbers give full details.     |    back to list of philosophers     |     unexpand these ideas

Ideas of Laurence Bonjour, by Text

[American, b.1943, Professor at the University of Washington, at Seattle.]

1980 Externalist Theories of Empirical Knowledge
Intro p.10 Externalist theories of knowledge are one species of foundationalism
     Full Idea: Externalist theories of knowledge are one species of foundationalism.
     From: Laurence Bonjour (Externalist Theories of Empirical Knowledge [1980], Intro)
     A reaction: I don't see why there shouldn't be a phenomenalist, anti-realist version of externalism, which just has 'starting points' instead of a serious commitment to foundations.
§5 p.27 The Lottery Paradox says each ticket is likely to lose, so there probably won't be a winner
     Full Idea: The Lottery Paradox says that for 100 tickets and one winner, each ticket has a .99 likelihood of defeat, so they are all likely to lose, so there is unlikely to be a winner.
     From: report of Laurence Bonjour (Externalist Theories of Empirical Knowledge [1980], §5) by PG - Db (ideas)
     A reaction: The problem seems to be viewing each ticket in isolation. If I buy two tickets, I increase my chances of winning.
§I p.11 The big problem for foundationalism is to explain how basic beliefs are possible
     Full Idea: The fundamental question that must be answered by any acceptable version of foundationalism is: how are basic beliefs possible?
     From: Laurence Bonjour (Externalist Theories of Empirical Knowledge [1980], §I)
     A reaction: This question seems to be asking for a justification for basic beliefs, which smacks of 'Who made God?' Look, basic beliefs are just basic, right?
§I p.11 The main argument for foundationalism is that all other theories involve a regress leading to scepticism
     Full Idea: The central argument for foundationalism is simply that all other possible outcomes of the regress of justifications lead inexorably to scepticism.
     From: Laurence Bonjour (Externalist Theories of Empirical Knowledge [1980], §I)
     A reaction: If you prefer coherence to foundations, you need the security of reason to assess the coherence (which seems to be an internal foundation!).
§II p.15 Extreme externalism says no more justification is required than the truth of the belief
     Full Idea: The most extreme version of externalism would be one that held that the external condition required for justification is simply the truth of the belief in question.
     From: Laurence Bonjour (Externalist Theories of Empirical Knowledge [1980], §II)
     A reaction: The question is, why should we demand any more than this? The problem case is, traditionally, the lucky guess, but naturalist may say that these just don't occur with any regularity. We only get beliefs right because they are true.
§IV p.20 External reliability is not enough, if the internal state of the believer is known to be irrational
     Full Idea: External or objective reliability is not enough to offset subjective irrationality (such as unexplained clairvoyance).
     From: Laurence Bonjour (Externalist Theories of Empirical Knowledge [1980], §IV)
     A reaction: A good argument. Where do animals fit into this? If your clairvoyance kept working, in the end you might concede that you 'knew', even though you were baffled about how you managed it.
§IV p.20 Even if there is no obvious irrationality, it may be irrational to base knowledge entirely on external criteria
     Full Idea: It may be that where there are no positive grounds for a charge of irrationality, the acceptance of a belief with only external justification is still subjectively irrational in a sense that rules out its being epistemologically justified.
     From: Laurence Bonjour (Externalist Theories of Empirical Knowledge [1980], §IV)
     A reaction: A key objection. Surely rational behaviour requires a judgement to be made before a belief is accepted? If you are consistently clairvoyant, you must ask why.
1985 The Structure of Empirical Knowledge
5.1 p.88 A coherence theory of justification can combine with a correspondence theory of truth
     Full Idea: There is no manifest absurdity in combining a coherence theory of justification with a correspondence theory of truth.
     From: Laurence Bonjour (The Structure of Empirical Knowledge [1985], 5.1)
     A reaction: His point is to sharply (and correctly) distinguish coherent justification from a coherence theory of truth. Personally I would recommend talking of a 'robust' theory of truth, without tricky commitment to 'correspondence' between very dissimilar things.
5.3 p.99 Anomalies challenge the claim that the basic explanations are actually basic
     Full Idea: The distinctive significance of anomalies lies in the fact that they undermine the claim of the allegedly basic explanatory principles to be genuinely basic.
     From: Laurence Bonjour (The Structure of Empirical Knowledge [1985], 5.3)
     A reaction: This seems plausible, suggesting that (rather than an anomaly flatly 'falsifying' a theory) an anomaly may just demand a restructuring or reconceptualising of the theory.
5.5 p.107 There will always be a vast number of equally coherent but rival systems
     Full Idea: On any plausible conception of coherence, there will always be many, probably infinitely many, different and incompatible systems of belief which are equally coherent.
     From: Laurence Bonjour (The Structure of Empirical Knowledge [1985], 5.5)
     A reaction: If 'infinitely many' theories are allowed, that blocks the coherentist hope that widening and precisifying the system will narrow down the options and offer some verisimilitude. If we stick to current English expression, that should keep them finite.
5.5 p.107 The objection that a negated system is equally coherent assume that coherence is consistency
     Full Idea: Sometimes it is said that if one has an appropriately coherent system, an alternative system can be produced simply be negating all of the components of the first system. This would only be so if coherence amounted simply to consistency.
     From: Laurence Bonjour (The Structure of Empirical Knowledge [1985], 5.5)
     A reaction: I associate Russell with this original objection to coherentism. I formerly took this to be a serious problem, and am now relieved to see that it clearly isn't.
5.5 p.107 A well written novel cannot possibly match a real belief system for coherence
     Full Idea: It is not even minimally plausible that a well written novel ...would have the degree of coherence required to be a serious alternative to anyone's actual system of beliefs.
     From: Laurence Bonjour (The Structure of Empirical Knowledge [1985], 5.5)
     A reaction: This seems correct. 'Bleak House' is wonderfully consistent, but its elements are entirely verbal, and nothing occupies the space between the facts that are described. And Lady Dedlock is not in Debrett. I think this kills a standard objection.
7.1 p.141 Empirical coherence must attribute reliability to spontaneous experience
     Full Idea: An empirical coherence theory needs, for the beliefs of a cognitive system to be even candidates for empirical justification, that the system must contain laws attributing a high degree of reliability to a variety of spontaneous cognitive beliefs.
     From: Laurence Bonjour (The Structure of Empirical Knowledge [1985], 7.1)
     A reaction: Wanting such a 'law' seems optimistic, and not in the spirit of true coherentism, which can individually evaluate each experiential belief. I'm not sure Bonjour's Observation Requirement is needed, since it is incoherent to neglect observations.
7.2 p.147 A coherent system can be justified with initial beliefs lacking all credibility
     Full Idea: It is simply not necessary in order for [the coherence] view to yield justification to suppose that cognitively spontaneous beliefs have some degree of initial or independent credibility.
     From: Laurence Bonjour (The Structure of Empirical Knowledge [1985], 7.2)
     A reaction: This is thoroughly and rather persuasively criticised by Erik Olson. But he always focuses on the coherence of a 'system' with multiple beliefs. I take the credibility of each individual belief to need coherent assessment against a full background.
8.3 p.171 The best explanation of coherent observations is they are caused by and correspond to reality
     Full Idea: The best explanation for a stable system of beliefs which rely on observation is that the beliefs are caused by what they depict, and the system roughly corresponds to the independent reality it describes.
     From: Laurence Bonjour (The Structure of Empirical Knowledge [1985], 8.3)
     A reaction: [compressed] Anyone who links best explanation to coherence (and to induction) warms the cockles of my heart. Erik Olson offers a critique, but doesn't convince me. The alternative is to find a better explanation (than reality), or give up.
1998 In Defence of Pure Reason
Pref p.-4 Philosophy is a priori if it is anything
     Full Idea: My conviction is that philosophy is a priori if it is anything.
     From: Laurence Bonjour (In Defence of Pure Reason [1998], Pref)
     A reaction: How about knowledge of a posteriori necessities, such as the length of a metre, known by observation of the standard metre in Paris?
§1.2 p.10 A priori justification requires understanding but no experience
     Full Idea: A proposition will count as being justified a priori as long as no appeal to experience is needed for the proposition to be justified - once it is understood.
     From: Laurence Bonjour (In Defence of Pure Reason [1998], §1.2)
     A reaction: Could you 'understand' that a square cannot be circular without appeal to experience? I'm losing faith in the pure a priori.
§1.3 p.12 The concept of possibility is prior to that of necessity
     Full Idea: While necessity and possibility are interdefinable concepts, it is the idea of a possible world or situation which is intuitively primary.
     From: Laurence Bonjour (In Defence of Pure Reason [1998], §1.3)
§3.5 p.79 Indeterminacy of translation is actually indeterminacy of meaning and belief
     Full Idea: The thesis of the indeterminacy of translation would be better described as the thesis of the indeterminacy of meaning and belief.
     From: Laurence Bonjour (In Defence of Pure Reason [1998], §3.5)
     A reaction: Not necessarily. It is not incoherent to believe that the target people have a coherent and stable system of meaning and belief, but finding its translation indeterminate because it is holistic, and rooted in a way of life.
§3.6 p.84 The induction problem blocks any attempted proof of physical statements
     Full Idea: The attempt to prove physical statements on the basis of sensory evidence is defeated by the problem of induction.
     From: Laurence Bonjour (In Defence of Pure Reason [1998], §3.6)
     A reaction: This sounds like a logician's use of the word 'prove', which would be a pretty forlorn hope. Insofar as experience proves anything, fully sensing a chair proves its existence.
§3.7 p.96 Externalism means we have no reason to believe, which is strong scepticism
     Full Idea: If externalism is the final story, we have no reason to think that any of our beliefs are true, which amounts to a very strong and intuitively implausible version of scepticism.
     From: Laurence Bonjour (In Defence of Pure Reason [1998], §3.7)
     A reaction: A very good point. I may, like a cat, know many things, with good external support, but as soon as I ask sceptical questions, I sink without trace if I lack internal reasons.
§3.7 p.96 Externalist theories of justification don't require believers to have reasons for their beliefs
     Full Idea: An externalist theory of epistemic justification or warrant need not involve the possession by the believer of anything like a reason for thinking that their belief is true.
     From: Laurence Bonjour (In Defence of Pure Reason [1998], §3.7)
     A reaction: That is the problem with externalism. If the believer does not have a reason, then why would they believe? Externalists are interesting on justification, but daft about belief. Why do I believe I know something, when I can't recall how I learnt it?
§3.7 n50 p.92 Coherence can't be validated by appeal to coherence
     Full Idea: The epistemic authority of coherence cannot itself be established by appeal to coherence.
     From: Laurence Bonjour (In Defence of Pure Reason [1998], §3.7 n50)
     A reaction: The standard approach amongs modern philosophers (following, I think, Kripke) is to insist on 'intuition' as basic, despite all its problems. I have no better suggestion.
§4.1 p.98 You can't explain away a priori justification as analyticity, and you can't totally give it up
     Full Idea: Moderate empiricists try unsuccessfully to explain a priori justification by means of analyticity, and radical empiricist attempts to dispense with a priori justification end in nearly total scepticism.
     From: Laurence Bonjour (In Defence of Pure Reason [1998], §4.1)
     A reaction: My working theory is neither of the above. Because we can abstract from the physical world, we can directly see/experience generalised (and even necessary) truths about it.
§4.1 p.99 Moderate rationalists believe in fallible a priori justification
     Full Idea: Moderate rationalism preserves a priori justification, but rejects the idea that it is infallible.
     From: Laurence Bonjour (In Defence of Pure Reason [1998], §4.1)
§4.3 p.110 Perceiving necessary connections is the essence of reasoning
     Full Idea: If one never in fact grasps any necessary connections between anything, it is hard to see what reasoning could possible amount to.
     From: Laurence Bonjour (In Defence of Pure Reason [1998], §4.3)
§4.5 p.119 A priori justification can vary in degree
     Full Idea: A priori justification can vary in degree.
     From: Laurence Bonjour (In Defence of Pure Reason [1998], §4.5)
     A reaction: This idea, which I trace back at least to Russell, seems to me one of breakthrough ideas in modern thought. It means that a priori knowledge can be reconnected with a posteriori knowledge.
§5.2 p.131 Our rules of thought can only be judged by pure rational insight
     Full Idea: Criteria or rules do not somehow apply to themselves. They must be judged by the sort of rational insight or intuition that the rationalist is advocating.
     From: Laurence Bonjour (In Defence of Pure Reason [1998], §5.2)
§6.7 p.180 All thought represents either properties or indexicals
     Full Idea: I assume that the contents of thought can be accounted for by appeal to just two general sorts of ingredient - properties (including relations) and indexicals.
     From: Laurence Bonjour (In Defence of Pure Reason [1998], §6.7)
     A reaction: I don't accept that relations are a type of properties. Since he does not include objects or substances, I take it that he considers objects to be bundles of properties.
§7.7 p.214 Induction must go beyond the evidence, in order to explain why the evidence occurred
     Full Idea: Inductive explanations must be conceived of as something stronger than mere Humean constant conjunction; …anything less than this will not explain why the inductive evidence occurred in the first place.
     From: Laurence Bonjour (In Defence of Pure Reason [1998], §7.7)
2003 A Version of Internalist Foundationalism
1.4 p.16 It is hard to give the concept of 'self-evident' a clear and defensible characterization
     Full Idea: Foundationalists find it difficult to attach a clear and defensible content to the idea that basic beliefs that are characterized as 'self-justified' or 'self-evident'.
     From: Laurence Bonjour (A Version of Internalist Foundationalism [2003], 1.4)
     A reaction: A little surprising from a fan of a priori foundations, especially given that 'self-evident' is common usage, and not just philosophers' jargon. I think we can talk of self-evidence without a precise definition. We talk of an 'ocean' without trouble.
1.5 p.21 The concept of knowledge is so confused that it is best avoided
     Full Idea: The concept of knowledge is seriously problematic in more than one way, and is best avoided as far as possible in sober epistemological discussion.
     From: Laurence Bonjour (A Version of Internalist Foundationalism [2003], 1.5)
     A reaction: Two sorts of states seem to be conflated: one where an animal has a true belief caused by an environmental event, and the other where a scholar pores over books and experiments to arrive at a hard-won truth. I say only the second is 'knowledge'.
2.1 p.25 Reliabilists disagree over whether some further requirement is needed to produce knowledge
     Full Idea: Reliabilist views differ among themselves with regard to whether a belief's being produced in a reliable way is by itself sufficient for epistemic justification or whether there are further requirements that must be satisfied as well.
     From: Laurence Bonjour (A Version of Internalist Foundationalism [2003], 2.1)
     A reaction: If 'further requirements' are needed, the crucial question would be which one is trumps when they clash. If the further requirements can correct the reliable source, then it cannot any longer be called 'reliabilism'. It's Further-requirement-ism.
2.2 p.27 If the reliable facts producing a belief are unknown to me, my belief is not rational or responsible
     Full Idea: How can the fact that a belief is reliably produced make my acceptance of that belief rational and responsible when that fact itself is entirely unavailable to me?
     From: Laurence Bonjour (A Version of Internalist Foundationalism [2003], 2.2)
     A reaction: This question must rival Pollock's proposal (Idea 8815) as the master argument against externalism. Bonjour is assuming that knowledge has to be 'rational and responsible', but clearly externalists take a more lax view of knowledge.
3.1 p.45 My incoherent beliefs about art should not undermine my very coherent beliefs about physics
     Full Idea: If coherentism is construed as involving the believer's entire body of beliefs, that would imply, most implausibly, that the justification of a belief in one area (physics) could be undermined by serious incoherence in another area (art history).
     From: Laurence Bonjour (A Version of Internalist Foundationalism [2003], 3.1)
     A reaction: Bonjour suggests that a moderated coherentism is needed to avoid this rather serious problem. It is hard to see how a precise specification could be given of 'areas' and 'local coherence'. An idiot about art would inspire little confidence on physics.
3.2 p.53 Coherence seems to justify empirical beliefs about externals when there is no external input
     Full Idea: The 1st standard objection to coherence is the 'isolation problem', that contingent apparently-empirical beliefs might be justified in the absence of any informational input from the extra-conceptual world they attempt to describe.
     From: Laurence Bonjour (A Version of Internalist Foundationalism [2003], 3.2)
     A reaction: False beliefs can be well justified. In a perfect virtual reality we would believe our experiences precisely because they were so coherent. Messengers from the front line have top priority, but how do you detect infiltrators and liars?
3.2 p.54 Coherentists must give a reason why coherent justification is likely to lead to the truth
     Full Idea: The 3rd standard objection to coherence is the demand for a meta-justification for coherence, a reason for thinking that justification on the basis of the coherentist view of justification is in fact likely to lead to believing the truth.
     From: Laurence Bonjour (A Version of Internalist Foundationalism [2003], 3.2)
     A reaction: Some coherentists respond by adopting a coherence theory of truth, which strikes me as extremely unwise. There must be an underlying optimistic view, centred on the principle of sufficient reason, that reality itself is coherent. I like Idea 8618.
3.2 p.54 For any given area, there seem to be a huge number of possible coherent systems of beliefs
     Full Idea: The 2nd standard objection to coherence is 'alternative coherent systems' - that there will be indefinitely many possible systems of belief in relation to any given subject area, each as internally coherent as the others.
     From: Laurence Bonjour (A Version of Internalist Foundationalism [2003], 3.2)
     A reaction: This seems to imply that you could just invent an explanation, as long as it was coherent, but presumably good coherence is highly sensitive to the actual evidence. Bonjour observes that many of these systems would not survive over time.
4.2 p.67 If neither the first-level nor the second-level is itself conscious, there seems to be no consciousness present
     Full Idea: In the higher-order thought theory of consciousness, if the first-order thought is not itself conscious and the second-order thought is not itself conscious, then there seems to be no consciousness of the first-level content present at all.
     From: Laurence Bonjour (A Version of Internalist Foundationalism [2003], 4.2)
     A reaction: A nice basic question. The only plausible answer seems to be that consciousness arises out of the combination of levels. Otherwise one of the levels is redundant, or we are facing a regress.
4.3 p.73 Conscious states have built-in awareness of content, so we know if a conceptual description of it is correct
     Full Idea: If we describe a non-conceptual conscious state, we are aware of its character via the constitutive or 'built-in' awareness of content without need for a conceptual description, and so recognise that a conceptually formulated belief about it is correct.
     From: Laurence Bonjour (A Version of Internalist Foundationalism [2003], 4.3)
     A reaction: This is Bonjour working very hard to find an account of primitive sense experiences which will enable them to function as 'basic beliefs' for foundations, without being too thin to do anything, or too thick to be basic. I'm not convinced.
5.1 n3 p.78 The adverbial account will still be needed when a mind apprehends its sense-data
     Full Idea: The adverbial account of the content of experience is almost certainly correct, because no account can be given of the relation between sense-data and the apprehending mind that is independent of the adverbial theory.
     From: Laurence Bonjour (A Version of Internalist Foundationalism [2003], 5.1 n3)
     A reaction: This boils down to the usual objection to sense-data, which is 'cut out the middle man'. Bonjour is right that at some point the mind has finally to experience whatever is coming in, and it must experience it in a particular way.