Modern anti-realists, of the semantic variety, maintain that the notion of truth contained within the traditional truth-conditional account of meaning must be understood as being evidentially constrained. They argue that we are only guaranteed to understand those statements for which there is evidence available, and thus that we are only capable of ‘grasping truth-conditions’ if this is understood in an anti-realistically acceptable manner. Often this is put in terms of verification: we are only guaranteed to understand those statements that are, in some sense, verifiable. Why in some sense? Well, there are three dimensions to this. First, some think that someone has to actually be in possession of the verifying evidence, whilst others think the availability of evidence is enough. Secondly, where the second account is taken, there are various conceptions of what notion of availability for the verifying evidence is the correct notion to focus on – and these are just some of the options roughly characterised –‘actually verifiable by some actual person now alive’, ‘actually verifiable by some person who was alive at some point’, verifiable by some actual person in principle', ‘would have been verifiable by some person if they had existed in the right place at the right time’… Third, there is disagreement over whether ‘verification’ is itself the right notion to utilise in the formulation of anti-realism. Some say the correct notion if warranted assertibility, some say justification, others say superassertibility.1 But, it does not matter here which is correct, for it is possible to subsume all of these accounts under two general schemas (where CR stands for some suitable cognitive relation such as being warranted in asserting that):
ANT1: Necessarily, a proposition P is true iff some agent A stands in CR to P.
ANT2: Necessarily, a proposition P is true iff some agent A (either possible or actual) would stand in CR to P if they were appropriately situated and were in possession of the available evidence.2
Any account that falls under either one of these general schemas is an anti-realist account. Of course, Berkeley is often though of as being the original anti-realist. So the first question that I want to address here is whether Berkeley's account can be plausibly presented as falling under one of these schemas. I think that it can.
II. BERKELEY'S ACCOUNT IN ANTI-FORM
Berkeley, of course, did not talk in terms of propositions. Instead his account was formulated directly in terms of the existence of objects qua Ideas. That is, he accepted an Ideational theory of the mind and, because he maintained that the notion of matter was incoherent, he thought that only Ideas (thought of, more or less, as mental objects) exist. He furthermore maintained that Ideas could not exist outside of any mind, and he equated perception with the having of Ideas. Thus, an agent perceives if and only if they have Ideas, only Ideas exist, and so esse est percipi and percipi est esse. As such, he would have accepted the following two propositions:
A1: Necessarily, if an object O exists, then there is some agent A that has an Idea I of O.
A2: Necessarily, if some agent A has an Idea I of an object O, then O exists.
But we should be careful how we read these propositions. The natural reading is to take the agent-quantifiers as ranging over all and only human agents.3 But whilst Berkeley would have accepted this reading of A2, he would not have accepted it of A1. In fact, he would only have accepted A1 read with the agent-quantifiers ranging over omniscient non-human agents. As he thought, in addition, that there is only one omniscient non-human agent (i.e. God)4 we can give a more perspicuous formulation of Berkeley's Idealism as being constituted by the acceptance of the following propositions:
A1*: Necessarily, if an object O exists, then there is some omniscient non-human agent A that has an Idea I of O.
A2*: Necessarily, if some non-human omniscient agent A has an Idea I of an object O, then O exists.
Whilst this formulation is not given in terms of propositions, it can be represented as such. For an agent A to have an Idea I of O is certainly for that agent to be in some cognitive state. Given that an object O exists iff the proposition ‘O exists’ is true5, then we can say that an agent A has an Idea I of O iff they fall into the having an Idea that relation to the proposition ‘O exists’. Thus we get:
And this is clearly of form ANT1, albeit restricted to propositions about the existence of objects. This restriction does nothing to temper Berkeley's anti-realism however. An anti-realist account that applies to all propositions about the existence of things in the world is a strong anti-realist account.
In what follows I will offer a reading of Berkeley that: 1. makes it clear that this is an acceptable way of formulating Berkeleyean Idealism, 2. offers a revealing insight into the reasons why Berkeley accepted this position, and 3. allows us to see that these reasons are, in fact, bad ones.
III. WHY DID BERKELEY ACCEPT A*?
The answer, it seems to me, is that Berkeley felt two distinct pressures.
The first pressure that Berkeley felt is constituted by the fact that many of the Ideas that we (human agents) have occur in our minds unbidden. If one is impressed by this fact, as Berkeley was, it is natural to ask: what causes these ideas to occur? As we shall see, Berkeley thought that he had to answer this question by appealing to God.
Above I said that Berkeley equated perception with the having of Ideas. But this is not an entirely satisfactory way of putting things. It is true that for Berkeley every Idea is as real as every other, and all instances of having Ideas count as perceivings. But Berkeley's use of the term ‘perception’ covers all experiential occurrences, including those that we would not normally count as perceivings (such as imaginings). Despite this, Berkeley was aware that not all instances of the having of Ideas should be counted as genuine instances of perception (i.e. what we would call simply ‘perception’). He thus recognised the need to distinguish between those experiences that do count as genuine perceptions and those that do not. This constitutes the second pressure.
There are intrinsic experiential properties that our experiences possess that offer evidence to us that our experiences are genuine cases of direct perception and not illusion. There are those features that Hume points to in order to characterise what ‘impressions’ are, viz. that our experiences are forceful, lively and vivacious. And there are other hard-to-spell-out features to do with lucidity – that our current experiences conform with our past experiences, that objects appear to behave as we expect them to behave (e.g. they do not appear to flit in and our of existence and they appear to follow spatio-temporal paths), and so on. For short, when our experiences do have all of these features let us say that they are well-structured. It is perfectly possible for a person to have well-structured hallucinations. But Berkeley, in considering how to make the distinction between genuine and non-genuine perception, nowhere considers this possibility. Instead he focuses almost solely on the distinction between genuine perceptions and imaginings: he recognised the fact that it is possible to create Ideas in the imagination, but did not want to count the having of such Ideas as genuine cases of perception. The following remarks are apposite here: if one focuses, as Berkeley does, solely upon the distinction between imaginings and all other experiential occurrences when attempting to draw the distinction between genuine and non-genuine perception, the obvious point to note is that whilst imaginings are subject to the will, all other experiential occurrences are not. Moreover, one this distinction is drawn, one is likely to be struck by the fact that all (or nearly all) of the unbidden experiences that we have are, as a matter of fact, well-structured. Both of these considerations, I think, play a part in Berkeley's thinking. The first point, that those experiences that are not imaginings occur unbidden, is one that Berkeley certainly sought to exploit. But this is where the second pressure combines with the first: he argues further than where our Ideas are not caused by our own will, they must be caused by some other will:
But whatever power I may have over my own thoughts, I find the Ideas actually perceived by sense have not a like dependence on my own will. When in broad day-light I open my eyes, it is not in my power to choose whether I shall see or no, or to determine what particular objects shall present themselves to my view; and so likewise as to hearing and the other senses, the ideas presented on them are not creatures of my will. There is therefore some other will or spirit that produces them. [Emphasis added]6
The will, or spirit, that Berkeley thinks causes our unbidden Ideas is, of course, God. Thus the very fact that we have Ideas that are not caused by our own volition functions in Berkeley's account as a proof for the existence of God.7 There are a number of distinct worries about this proof.8 But the fundamental one is that the argument is based upon the dubious premise that if our Ideas are not caused by our own will, then they must be caused by some other will. The pertinent question is: Why does Berkeley think that this dubious premise is correct? We shall return to this question very shortly. But first let us deal with how Berkeley did draw the distinction between genuine and non-genuine cases of perception, for this will give rise to a second pertinent question.
There is some reason for thinking that Berkeley tried to make the distinction between genuine and non-genuine perception by a direct appeal to the well-structuredness of our Ideas themselves. Indeed, it is this interpretation that Jonathan Bennett favours. He says:
Wanting to know whether a certain Idea is (of) a real thing, I ask Berkeley for help. He tells me to discover whether the idea came to me unbidden, and whether it is ‘strong’ (he also says ‘vivid’, ‘distinct’, ‘lively’); and I report that it passes both of these tests. Berkeley rightly thinks that this is not enough: ideas count as (ideas of) real things, he says, only if they are also ‘orderly and coherent’, ‘regular’, ‘constant’, ‘steady’, ‘not excited at random’, ‘connected’.9
But I do not see this interpretation of Berkeley as correct. For sure, Berkeley did maintain that those Ideas that are well-structured do invariably count as cases of genuine perception. But this, I think, is because Berkeley was struck by the fact mentioned earlier, that all of our unbidden experienced are, as a matter of fact, well-structured. But just because Berkeley was struck by this fact does not mean that he wished to present it as an account of what direct perception consists in. Consider the following passage from the Principles that Bennett takes many of the quoted phrases above from:
The ideas of sense are more strong, lively, and distinct than those of imagination; they have likewise a steadiness, order and coherence, and are not excited at random, as those which are the effects of human will often are, but in a regular train or series, the admiral connexion whereof sufficiently testifies the wisdom and benevolence of its Author. [Emphasis added]10
Here Berkeley merely reports that cases of genuine perception (‘ideas of sense’) are well-structured. He does not claim that they must be. But what he does make clear is that he thinks that our genuine perceptions are well-structured because they have been caused in us by God. Because Berkeley does not consider the possibility of well-structured hallucination or illusion I think that he takes the occurrence of well-structured experiences as concrete evidence that genuine perception is occurring. But I do not think that this constitutes, for him, a criterion for the occurrence of genuine perception as Bennett thinks. Instead, the criterion for genuine perception is that the Ideas themselves have been caused in us by God. That this is the correct interpretation is supported by the following striking sentence that occurs just a few paragraphs later in the Principles:
The ideas imprinted on the senses by the Author of Nature are called real things.11
Thus I think that Berkeley's definition of genuine perception comes out as something like the following:
A human agent A genuinely perceives an object O iff A has an Idea I of O and A's having of I is caused by God.
Now, Berkeley thought that the Ideas present in my mind when genuine perception occurs also have to be present in the mind of God. But this does not follow from the above criterion. All that follows is that God has to be the cause of those Ideas. So here is the second pertinent question: why does Berkeley think that they do have to be present in the mind of God?
So, we have two questions: 1. Why does Berkeley think that if our Ideas are not caused by our own will, then they must be caused by some other will? 2. Why does Berkeley think that the Ideas caused to be in my mind by God also have to be in God's mind? There is a line in Bennett that enables us to answer both of these questions (although Bennett does not utilise the line in the way that I do.) Bennett makes the following hypothesis:
Berkeley frequently uses ‘depend’ and its cognates to express relations between ideas and minds or spirits; in some of these uses ‘I depends on S’ means ‘I is had by S’, in others it means ‘I is caused by S’; and Berkeley is never aware of this ambiguity.12
I am aware that this hypothesis has been rejected by some. But Bennett does present substantial textual evidence for this hypothesis, and by showing how the hypothesis can be made to do new work, I intend the following to constitute a defence of it. What is the new work? Well, simply that it makes Berkeley's acceptance of the following argument perfectly intelligible:
- (a)If any idea comes into my mind unbidden, then it does not depend upon (=is not caused by) me.
- (b)All ideas depend upon (=are had by) some being.
- (c)There are idea that come into my mind unbidden (=are not caused by me).
- (d)Those ideas must depend upon (=must be had by and be caused by) some other being (i.e. God),
That Berkeley did reach conclusion (d) by just this (fallacious) route fits perfectly with the text.
It is now a short step to establishing that Berkeley would have accepted A* and would have rejected A1 (read with the quantifiers ranging over all and only human agents). Berkeley understood the above account as establishing that Ideas that occur in human minds are genuine perception (and therefore that the objects O that they are of exist) only if and because they also occur in the mind of God. Thus we establish A2*.13 But, moreover, if this is so, and if God is an independently existing being, then there is nothing preventing Ideas occurring in the mind of God even if they do not occur in any human mind:
[A]ll the choir of heaven and furniture of the earth … have not any subsistence, without a mind, that there being is to be perceived or known; that consequently so long as they are not actually perceived by me, or do not exist in my mind or that of any other created spirit, they must either have no existence at all, or else subsist in the mind of some eternal spirit. [Emphasis added]14
So no object can exist unless God perceives it, but there can be objects that exist unperceived by any human agent. That is:
Not necessarily, it is not the case that there are objects O such that no human agent A has an Idea I of O.
So A1 read as specified is false. But:
Necessarily, for all objects O, God has an Idea I of O.
So A1* is true, and combining with A2* we get A*.