‘Or’ or ‘or’

An unimportant note; but the opportunity to use this title was, naturally, irresistible.

Discussion with F2 today usefully exposed the fact that the word ‘or’ is ambiguous in its usage.

  • Sometimes, when we say ‘P or Q’, we mean ‘either P or Q, but not both‘.
  • Sometimes, when we say ‘P or Q’, we mean ‘P or Q, or both P and Q‘.

Call the first of these usages the restrictive, or non-inclusive, use of ‘or’.

Call the second of these usages the non-restrictive, or inclusive, use of ‘or’.

Gettier, Case II

The distinction arose in discussion of Gettier’s Case II.

Remember that a ‘Gettier example’ is a case designed to challenge the sufficiency of the tripartite theory of knowledge. It is an imagined case in which a person satisfies all three conditions in the tripartite theory (P is true, (s)he believes that P, and (s)he is justified in believing that P); but in which it seems wrong to say that the person knows the relevant proposition (P). Both cases in Gettier’s 1963 paper involve the subject’s forming a justified and true belief by deducing (= validly inferring) it from a justified but false belief. In Gettier’s Case II, the situation is that:

Smith has the justified (but false) belief that Jones owns a Ford.

Smith infers, quite correctly, and comes to believe, the following (disjunctive) proposition (call it P):

P: Jones owns a Ford or Brown is in Barcelona.

P turns out to be true because Brown is as a matter of fact in Barcelona, even though Smith does not know the truth of this disjunct. Thus: Smith is justified in believing that P (because P is inferred from another justified belief – that Jones owns a Ford); P is true; but Smith (intuition tells us) should not be counted as knowing that P. (After all, what makes P true is that Brown is in Barcelona; but Smith has no knowledge of Brown’s whereabouts; he believes that P only because he has the false belief that the first disjunct – that Jones owns a Ford – is true.)


What sense of ‘or’ is Gettier assuming? The restrictive or the non-restrictive one?

The answer must be that Gettier is assuming a non-restrictive ‘or’.


Because P above is thought of as validly deduced or inferred from the proposition that Jones owns a Ford.


For clarity:

Let us use the word ‘or‘ to denote the non-restrictive (or inclusive) usage of ‘or’.

Let us use ‘or*‘ to denote the restrictive (or non-inclusive) usage of ‘or’.

Now: Gettier assumes that Smith can validly infer various disjunctive propositions from the proposition that Jones owns a Ford. For example, he says that Smith can validly infer all of the following:

Jones owns a Ford or Brown is in Boston

Jones owns a Ford or Brown is in Barcelona

Jones owns a Ford or Brown is in Brest-Litovsk

His idea must be that you can validly infer any proposition of the form ‘P or Q’ from ‘P'; it doesn’t matter what proposition Q is – the inference will still be valid.

But: Can you validly infer (P or* Q) from P, for no matter what proposition Q?

The correct answer is No: you can infer (P or* Q) from P only if the proposition Q is such that, if P is true, then Q is false.

This is because you can validly infer one proposition from another only if there is no possible situation in which the inferred proposition is false while the proposition from which it is inferred is true.

Thus if P is true, then (P or* Q) will be true only if Q is false. This means that you cannot infer (P or* Q) from P, regardless of what Q is.

Suppose, for example, that Q is some necessary truth, such as ‘bachelors are unmarried’. In this case, if P is true, Q will also be true; in which case (P or* Q) will be false.


You can validly infer (P or Q) from P, no matter what proposition Q is. For the disjunctive proposition (P or Q) is counted as true even if both P and Q are true. So there is no possible situation in which P is true but (P or Q) is false, no matter what proposition Q is.

Experiencing God… knowing God?

OK – another go at this, in case it helps.

To experience God is to know that He exists. Discuss.

Well, for starters, to experience an F is not necessarily to know that any F exists. Rationale:

1. Lack of concept

For example, I can experience a hurricane without knowing that hurricanes exist. Proof: If I lack the concept of a hurricane, I will inevitably lack the relevant knowledge. (I may know that something exists; but I don’t know that hurricanes do. Toss a baby into a hurricane; does it know that hurricanes exist? No. It lacks the relevant concept.)

So, minimally, I need a concept of the thing that I experience if my experience of it is to give me knowledge of its existence. This goes for God as much as for anything else.

2. Lack of recognition

Even if I possess the relevant concept, my experience of an F will not give me knowledge that an F exists unless I can recognise the experience as an experience of an F. So, for example, if God appears to me in disguise (as, say, Homeric gods frequently do) then I will be experiencing Him without necessarily knowing that He exists.

3. Lack of belief

Even if I possess the concept, and even if God appears to me undisguised, I will not thereby gain knowledge that God exists unless I believe that God exists as a result of my experience. This is because of the so-called ‘belief-condition’ of knowledge: knowing that p requires, as a minimum, the belief that p. So if I am pre-disposed to be hyper-sceptical about God, then I won’t know that God exists, even if He presents Himself to me without disguise, and even if I have no other reason to doubt the veridicality of the experience.

4. Presence of other factors undermining knowledge

Even if I possess the concept; and even if God appears to me undisguised; and even if as a matter of fact I believe that God exists on the basis of the experience, I cannot be said to know that God exists, on the basis of the experience, if I have substantial grounds for doubt as to its veridicality – either concerning the experience itself or independent grounds for doubting the existence of God.

This fourth point is the biggie: the ultimate problem is going to be that, no matter how ‘good’ the experience, in relation to the criteria to be established, there is bound to be enough doubt of one kind or another to undermine the claim that the experience gives one knowledge that God exists.

So much by way of intro; now let’s get to business:

Arguments from religious experience are supposed to justify belief in the existence of God on the basis of experience of Him.

Various sorts of experience have been claimed to do the job, ranging from direct experience of God of the kind described in the Old Testament (God’s speaking to Abraham and the like), through experiences of ordinary phenomena (such as a sunrise) under some religious aspect, to dreams and visions, experience of apparent miracles, and mere numinous feelings.

Such experiences are supposed to provide an independent and direct proof of God’s existence, of a non-inferential, direct, form that can be contrasted with the indirect and inferential form of arguments such as the Cosmological Argument and the Argument from Apparent Design.

The question we face is really the following: Suppose that I am the subject of one or another of these experiences; can I know, on the basis of such an experience, that God exists? The answer must be that I can know this only if I have good reason to take the experience at face value.

Why should you trust your experiences?

In general, I have reason to take my experiences at face value (as being veridical) only insofar as I can reasonably apply the Principle of Credulity. That principle is the rule that, other things being equal, I am entitled to believe that what I seem to perceive is true.

But what backs the Principle of Credulity?

Only, I think, the fact that perception is a generally veridical process – one that more often than not gives one correct information about the world. Perception, as an evolved faculty (or set of faculties), does have a good claim to do this: evolution would naturally tend to favour genetic mutations that gave us accurate, as opposed to inaccurate, information. There isn’t any good reason to suppose that one’s experiences quite generally are veridical: we know that we are subject to all sorts of hallucinations, dreams, fantasies, etc., that undermine any more general proposed rule that the content of all experience should in general be trusted.

It follows that, to the extent that there is reason to doubt that a religious experience is perceptual, I lack reason for applying the Principle of Credulity; and I (therefore) lack grounds for thinking that the experience gives me knowledge of the existence of God.

Even if I lack specific grounds for suspecting that the experience is of a non-perceptual nature, I may have grounds for suspecting that the content of the experience is non-veridical. In the same way, I may in come circumstances know that I am perceiving something; but if it is radically different from the way that I believe it to be, then I am likely to distrust the content of my experience.

Now let’s put all these materials together:

Some of the types of RE mentioned earlier seem distinctly unpromising as candidates for perceptual awareness of God.

(a) Consider ‘numinous feelings‘, for example. These don’t sound particularly like perceptual experiences at all – arguably they are more like emotions, or beliefs that arise independently of any specifically perceptual process.

(b) Dreams and ‘visions’ are little better: although such experiences are presented in the visual modality, we are familiar with the idea that there are non-perceptual experiences that are also so presented – such as hallucinations and ordinary dreams. So without a special reason to think that dreams and visions with a religious content are perceptual, we seem equally entitled to class them with the non-perceptual cases.

(c) Perception of apparent miracles is a slightly different case. Here, we are presumably perceiving something: we are perceiving (say) water being turned into wine; or a dead man standing up and walking; or whatever. So these cases cannot be so easily dismissed as non-perceptual as the former cases. The issue is whether we are perceiving anything that gives us reason to believe in God. But certainly the perception itself is not giving us knowledge of God directly, as the argument intended; at most, we may come to have reason to believe that God exists on the basis of some inference, in which we reason that the events apparently perceived could not have occurred without a divine cause. So it couldn’t be said that the perceptual experience itself gives us sufficient reason to believe in God, let alone knowledge that He exists.

(d) Perceiving some natural event (like a sunrise) as having a divine significance is different again: here,we are undoubtedly perceiving the sunrise; the real issue is whether the specifically divine aspect of the experience is also part of the perception, or whether instead it is something non-perceptual that we bring to the experience. And in such cases, the fact that the religious aspect of the experience is not typically shared by all observers of the perceived phenomenon is what gives us grounds for doubt that it is a genuine perceptual aspect.

So the case that we are left with – the best case for the argument – is the case in which we seem to perceptually confronted by God Himself. But even in this best possible case, there are likely to be reasonable doubts – for example:

(i) If the experience is provoked (or apparently provoked) by special circumstances (fasting etc), then it is bound to provoke suspicion, since we know that such special circumstances can cause hallucinations and the like.

(ii) If there are other potentially distorting factors, such as prior belief in God, wishful thinking, education, indoctrination and so on. We know that such factors can affect the content of our experience; to the extent that such factors are present, we should therefore be inclined to distrust our experiences.

(iii) If the experience is unique, or rare. If I’m alone (or nearly alone) in experiencing what I experience, the suspicion must arise that it is an experience that is due to some fault in me rather than an ordinary perceptual one.

(iv) If the content of the experience is unusual (which it is by definition bound to be, if it is an experience as of God), then there are going to be grounds for that reason alone to be suspicious of the view that it can be trusted. One generally assumes (and rightly so) that apparent perception as of unfamiliar and improbable-seeming things like unicorns is not to be trusted, but is rather to be suspected of being either illusory or hallucinatory.

Conclusion (roughly)

Even in the ‘best case’ scenario, there seem bound to be factors that undermine our entitlement to believe that experiences as of God can be trusted – either because they give us reason to doubt that the experiences are perceptual, or because they give us reason to suspect that, even if they are perceptual, they may not be veridical.

Even if, therefore, we assume that other background requirements are fulfilled (such as possession of the relevant concepts), it cannot be said that to experience God is to know that He exists. Even if we are on some occasion experiencing God, we are liable to have one or more of a host of reasons for doubting that that is what is going on; and all such reasons will undermine any claim to know that God exists on the basis of the experience.