Introduction
One of Berkeley's most direct arguments against materialism, and the representationalist epistemology which he takes the likes of Descartes and Locke to subscribe to, comes in §8 of the Principles. Footnote 1 There, Berkeley puts forward what scholars refer to as the ‘likeness principle’ (LP) which states that ‘an idea can be like nothing but an idea’. From LP, Berkeley argues that since the qualities which inhere in an unperceivable material substance could never resemble our ideas our ideas could not possibly represent those qualities.Footnote 2 If Berkeley's argument is successful, then it undermines the widely held view that our knowledge of external things in the sensible world comes via ideas in our minds which represent them. The likeness principle is one of the key premises in Berkeley’ argument in §8 of the Principles and there are several interpretations of Berkeley's treatment of it.Footnote 3 However, as several commentators have pointed out, another premise is required to reach Berkeley's conclusion that ideas could not possibly represent unperceivable qualities of material substances.Footnote 4 To establish that conclusion, Berkeley requires his readers to accept that ‘for one thing to represent another, those two things must resemble one another’. This premise is what we call the ‘resemblance thesis’ – and it appears to be similarly unargued for.
Various attempts have been made to explain why Berkeley accepts the resemblance thesis and why he fails to provide an explicit case for its defence. Most notably, Jonathan Hill has argued that Berkeley's acceptance of the resemblance thesis finds its roots in Cartesian discussions surrounding scepticism and intentionality. In what follows, we argue that there is a more local explanation – namely, the intellectual environment in Ireland and debates amongst Irish thinkers in which Berkeley was actively engaged. Our aim is not to undermine Hill's reading and we do not contest the claim that Cartesian thought influenced Berkeley's philosophy – on the contrary, Berkeley's De Motu clearly suggests that Berkeley was influenced by Cartesianism. We simply contend that, before looking far and wide for the source of Berkeley's views on representation and resemblance, one ought to begin this search locally; in Dublin and Ireland. Ultimately, we conclude that Berkeley's views on the relation between representation and resemblance were influenced by the immediate Irish context in which he was writing and that this has so far been underappreciated in the relevant scholarship.
We believe the resemblance thesis was more widespread in Early Modern Irish thought than we are able to demonstrate in this paper, but for the time being focus on tracing it through William Molyneux, William King, and Berkeley's engagement with both the Molyneux problem and debates in Ireland concerning the problem of divine attributes. Our contention is that Molyneux's problem of the ‘blind man made to see’, and early responses to it, make it evident that the resemblance thesis was an underlying and often implicitly accepted principle in Irish thought.Footnote 5 We thus aim to build on and substantiate David Berman's claim that the Molyneux problem is the ‘root metaphor’ of Early Modern Irish philosophy.Footnote 6
The structure of our argument is as follows. Firstly, we argue that early responses to Molyneux's problem of a blind man made to see established the resemblance thesis as an underlying principle in Early Modern Irish philosophy. The two earliest responses, from Molyneux himself and Locke,Footnote 7 emphasise the extent to which the objects of sight were seen to be unlike anything with which a blind man is familiar. In section two, we focus on William King's account of human knowledge of the divine attributes.Footnote 8 King draws on Molyneux's example, arguing we are no more familiar with God's attributes than a blind man is with the objects of sight. King also explicitly advocates an account of representation by means of likeness, thus suggesting that it is the failure of our knowledge of divine attributes to adequately resemble the true nature of those attributes that entails the former is only ever an inadequate representation. In section three, we consider Berkeley's response to King. Berkeley disagrees that our knowledge of divine attributes is comparable with a blind man's notion of the objects of sight, arguing that we use the knowledge we have of ourselves, as spirits, as an imperfect but nonetheless appropriate representation of the divine. Despite their disagreement, we emphasise that neither disputes the plausibility of the resemblance thesis itself. This suggests that both Berkeley and King found themselves working under the constraint of explaining how we can represent the divine attributes to ourselves by means of resemblance. Finally, we point to the fact that Berkeley uses Molyneux's original problem to defend his claim that the objects of touch and sight are fundamentally heterogenous (i.e., unalike) as further evidence that his acceptance of the resemblance thesis is rooted in Irish thought.
As we see it, there at least are two advantages to focusing on the resemblance thesis in Irish thought. Firstly, this discussion provides an insight into the more local influences on Berkeley's philosophy. There is no doubt that Berkeley took himself to be engaging with influential figures from across Europe including Descartes, Malebranche, and Locke. After all, the ‘materialism’ he opposes is, as he puts it, a ‘strangely [i.e. widely] prevailing’ view (PHK, §4). But the impact of other Irish thinkers on Berkeley's views should not be underestimated. Indeed, at least once Berkeley explicitly identifies himself as an Irish thinker.Footnote 9 Secondly, we see this discussion as beneficial to Early Modern scholarship more generally. By focusing on the local influences on Berkeley, a ‘canonical’ figure in Early Modern scholarship, this paper will shed light on Ireland's wider contribution to important epistemological debates concerning representation and knowledge via ideas.
1. The Resemblance Thesis
In this section, our aim is to establish the importance of the resemblance thesis in Berkeley's argument in §8 of the Principles and outline the interpretative problem that arises from his failure to justify or defend it. The resemblance thesis, as we previously stated, is the claim that ‘for one thing to represent another, those two things must resemble one another’. Before proceeding, it is worth clarifying our own construal of ‘representation’ in Berkeley's writing. Berkeley himself uses the term ‘represent’ in both a strict and a loose sense. In a strict sense, Berkeley understands representation as literal re-presentation of an object – such as when I imagine an idea which is an ‘image’ or ‘copy’ of something I have perceived via the senses (PHK, §33; see also PHK, §27). It is this re-presentation which requires resemblance and since Berkeley deems resemblance to be a necessary relation (NTV, §§45; TVV, §42–43 & §61), in what follows, when we refer to representation in a strict sense it should be taken to refer to a non-arbitrary relation. This non-arbitrariness is what distinguishes it from Berkeley's use of representation in a loose sense. When Berkeley uses the term ‘represent’ in a loose sense it is used synonymously with ‘signify’.Footnote 10 Signification relations are those shared between a sign and a thing signified and, importantly, in contrast with representation in a strict sense, they are arbitrary relations. In this paper, we are primarily concerned with Berkeley's employment of the term ‘represent’ in a strict sense as outlined above. By attributing the resemblance thesis to Berkeley, then, we take him to accept that representation, in this strict sense, requires resemblance.Footnote 11
At least two passages make it clear that Berkeley both accepts the resemblance thesis and takes his opponents to accept it too. Firstly, the following passage from the unpublished Manuscript Introduction to the Principles:
There is no similitude or resemblance betwixt words & the ideas that are marked by them. Any name may be used indifferently for the sign of any idea, or any number of ideas, it not being determin'd by any likeness to represent one more than another. But it is not so with ideas in respect of things, of which they are suppos'd to be the copies & images. They are not thought to represent them any otherwise, than as they resemble them. Whence it follows, that an idea is not capable of representing indifferently any thing or number of things it being limited by the likeness it beares to some particular existence, to represent it rather than any other. (MI, §12, our emphasis)
Berkeley claims the key difference between (i) words and their objects, and (ii) ideas and their objects is that a word can arbitrarily signify any idea, without resembling it, while ideas themselves can only represent that which they resemble.Footnote 12 Words, Berkeley explains, can signify ‘indifferently’. However, the representational capacity of an idea is ‘limited by the likeness it [bears]’. Even the terminology that Berkeley uses is important: the relation between words and ideas is one of ‘signification’, an arbitrary relation, whereas the relation between ideas and their objects is one of ‘representation’ which depends entirely upon what a certain idea does and does not resemble.Footnote 13 This passage is not repeated in Berkeley's published Introduction to the Principles, and there is no obvious counterpart in any of his published works.Footnote 14 Nonetheless, as Kenneth Winkler emphasises, that does not mean he ceased to believe that there is a difference between representation in a strict sense and signification, i.e. between a non-arbitrary and an arbitrary relation (Winkler, Reference Winkler1989, p. 12).
In §8 of the Principles, the resemblance thesis clearly informs both Berkeley's characterisation of his opponents’ view and his response to that view. There he writes:
But say you, though the ideas themselves do not exist without the mind, yet there may be things like them whereof they are copies or resemblances, which things exist without the mind in an unthinking substance. I answer, an idea can be like nothing but an idea; a colour or figure can be like nothing but another colour or figure. If we look but ever so little into our thoughts, we shall find it impossible for us to conceive a likeness except only between our ideas.
Hylas also puts forward a similar view in the Three Dialogues, maintaining that ‘real things or external objects [are] perceived by the mediation of ideas, which are their images and representations’ (DHP, 1.203). The first thing to note is that Berkeley presents his opponents (materialists who accept a representationalist epistemology) as holding the view that even if our ideas are the only things we immediately perceive – a claim which Berkeley takes himself to have established in the preceding sections – those ideas might simply be ‘copies or resemblances’ or ‘pictures or representations’ of qualities inhering in unthinking, material substances. The second thing to note is that Berkeley takes it that he can refute this position simply by emphasising that ideas can only resemble other ideas. Berkeley does not consider any other way in which an idea might represent its object beyond resemblance. While this might not be surprising in light of the view expressed in MI §12, it would be likely to surprise those of his opponents who held that the relation between ideas and material things is a causal one. As Jonathan Hill puts it:
[Berkeley] does not ever seem to consider the possibility that there might be material substances that do not resemble the objects of direct perception [i.e. ideas] at all, but which nonetheless cause them. (Hill, Reference Hill2011, p. 49)
It is also important to note that the kind of ‘representationalists’ that Berkeley is primarily concerned with also accept the ‘resemblance thesis’. Ideas, according to Berkeley's opponents, are ‘representations’ of ‘real things’ which Berkeley takes to be synonymous with saying they are ‘copies’, ‘images’, or ‘pictures’ of those things (PHK, §8; DHP, 1.203). It is clear, then, that according to Berkeley's reading of his opponents, ideas represent by means of resemblance. In Philonous’ words, according to representationalists, material objects are ‘represented or painted forth’ by ideas in our minds (DHP, 1.206).Footnote 15
John Carriero argues Berkeley's treatment of the relation between representation and resemblance is drawn from Medieval, Aristotelian accounts of sensible knowledge (Carriero, Reference Carriero2003). The Aristotelian view, broadly speaking, is that likeness is key to knowledge of the sensible world: ‘species’ that come to exist in the mind resemble what exists, and we get to know about what exists thanks to that resemblance relation. Berkeley's aim, on Carriero's reading, is to emphasise that without resemblance it is very difficult to consistently maintain we get to know what the world is really like. This is largely consistent with Kenneth Winkler's explanation of Berkeley's acceptance of the resemblance thesis. Without identifying any specific roots in scholastic thought, Winkler nonetheless focuses on establishing, for Berkeley, ideas which represent other objects do so in virtue of being images of those objects.Footnote 16
Such readings may help us to understand what Berkeley's view is, but they do little to explain Berkeley's treatment of the resemblance thesis in Principles §8 and why he (seemingly) did not feel the need to justify it. In that regard, a contextual reading is more helpful. Hill provides such a reading, charting the rise of the resemblance thesis after Descartes – for whom, ‘resemblance and causation cannot be separated’ – through later Cartesians such as Foucher, Du Hamel, and Malebranche. For all these thinkers, Hill argues, ‘The conclusion is evident […] representation must be about resemblance’ (Hill, Reference Hill2011, p. 56). Hill's claim is by taking Berkeley to be a part of this Cartesian tradition, and engaged in these debates, it is possible to explain why he never saw the need to justify his employment of the resemblance thesis.
Our aim is not to refute Hill's account. Nor do we think our explanation of Berkeley's acceptance of the resemblance thesis should displace Hill's. However, we do contend that our reading fills a gap in current literature surrounding this issue. It may well be the case that Berkeley also inherited the resemblance thesis from Cartesian thinkers and as far as he was engaging in their debates there was, thus, no need to explicitly justify its employment. However, as Hill's own discussion shows, even by the time Berkeley was writing, it was far from being a universally accepted axiom and Berkeley did not only engage himself in debates of the Cartesian tradition. Why, then, did Berkeley more generally take himself to be writing in an intellectual climate in which the resemblance thesis could be left unargued for – at least explicitly? Hill's reading offers one explanation as to why Berkeley saw it as unnecessary to defend resemblance in certain contexts. Our claim, however, is that it would be amiss not to also consider Berkeley's immediate, local context – in Ireland. We thus work on the reasonable assumption that Berkeley's immediate intellectual context is just as likely to have shaped his philosophical views as the wider context of eighteenth-century Europe. Our cause is strengthened by the fact that, throughout his career, many of Berkeley's philosophical views were explicitly and directly developed as responses to debates amongst Irish thinkers such as John Toland, William Molyneux, and William King.Footnote 17 This provides a compelling reason to search for the roots of Berkeley's resemblance thesis in Early Modern Irish philosophy. Thus, in what follows, we argue that Berkeley's acceptance of the resemblance thesis was likely to have been influenced by his engagement with William Molyneux, William King, and, more generally, thinkers in Ireland who were engaged in debates concerning representation and human knowledge of the divine attributes.
2. The Molyneux Problem
The aims of this section are to give an exposition of the Molyneux problem and to establish that the resemblance thesis is at least implicit in the two earliest responses to the problem; those of Molyneux himself and Locke. We thus trace the beginning of the ‘man born blind's’ journey through Early Modern Irish thought.
Molyneux first proposed the problem to Locke in a letter in 1688. There is no indication that Locke responded to this first letter and Molyneux sent another in 1693. Eventually, the problem was included in Locke's Essay from the second edition onwards. There, he quotes a section of Molyneux's letter verbatim, which reads:
Suppose a man born blind, and now adult, and taught by his touch to distinguish between a cube and a sphere […] Suppose then the cube and sphere placed on a table, and the blind man to be made to see; query; ‘Whether by his sight, before he touched them, he could now distinguish and tell, which is the globe, which the cube?’ (Essay II.ix.viii)Footnote 18
The problem Molyneux lays out is determining whether such a ‘blind man made to see’ would, based on his tactile experience of cubes and spheres, be able to tell the difference between a cube and sphere just by looking at them. Molyneux's answer is no. Locke agrees, citing the fact that the blind man made to see has no prior experience of how his visual experiences correspond with his tactile experiences. As he puts it, he has no experience ‘that what affects his touch so and so, must affect his sight so and so’. Locke adds to this that those who are not blind are ‘beholden to experience, improvement, and acquired notions’ (Essay II.ix.viii) in their ability to do what the Molyneux man cannot – namely, relate our visual experiences to our tactile ones (and vice versa). The more overt principle that underlies Molyneux and Locke's response to the problem, then, is that one needs to have experienced the correspondence between sensations of one kind and sensations of another to know that it exists.
Our contention is that there is an implicit principle at work here; namely, the resemblance thesis. For Molyneux and Locke, it is impossible for the blind man made to see to distinguish between a cube and a sphere by sight alone, because his representational capacities are restricted by likeness. The visual sensations which he begins to perceive as soon as he is ‘made to see’ do not represent anything to him, because they are not like anything he knows (up to that point). More specifically, the visual sensations with which he is now bombarded do not resemble any of the tactile sensations with which he has previously been acquainted. Prior to being made to see, cubes and spheres, for the blind man, are shapes that can be discerned by tactile sensation alone. In fact, it would be true to say that, for the blind man (before he is made to see), cubes and spheres just are things that are felt – assuming that shapes cannot be heard, smelt, or tasted. Without the ‘experience’ or ‘acquired notion’ that cubes and spheres can be seen as well as felt, it is just not possible, Molyneux and Locke take it, for the blind man to see cubes and spheres. The visual sensations that the blind man is ‘made to see’ represent nothing to him, prior to the experience that ‘what affects his touch so and so, must affect his sight so and so’, because they do not resemble anything with which he is currently familiar.
In this way, we contend that it is at least prima facie plausible that Molyneux and Locke's reactions to the ‘blind man made to see’ thought experiment were read in this way by subsequent Irish thinkers. This prima facie plausibility will suffice for our current purposes. As will become evident, we take it that the proof is in the pudding; by looking at how Molyneux's example came to be employed in subsequent Irish thought it becomes clear that it was taken to be a matter of representation and resemblance. Discussions concerning the relation of representation and resemblance would come to a head in the midst of a defining Irish debate; the issue of divine attributes. In the next section, we focus on William King's contribution to that debate, and argue that his own employment of the ‘Molyneux man’ indicates that he took it as illustrative of the resemblance thesis.
3. King and Divine Attributes
In this section, we demonstrate that the resemblance thesis plays a crucial role in the account of human knowledge of the divine attributes that William King puts forward in his Sermon. We also show that King's own employment of the example of a man born blind ties his own views, regarding representation and resemblance, to those of Molyneux and Locke. In this way, we begin to chart the progression of the resemblance thesis in Irish thought and the increasingly important role that the man born blind plays.
Despite the fact that the significance of William King in Irish history tends to go underappreciated, his influence within the context of Irish thought was quite significant.Footnote 19 He was not only the Archbishop of Dublin (1703-1729) and personally acquainted with both Molyneux and Peter Browne (Berkeley's provost at Trinity College), he was also a member of the Dublin Philosophical Society which Molyneux founded in 1683.Footnote 20 However, it is King's influence on Berkeley that is most significant for our current purposes. On March 1st, 1710, Berkeley wrote to his friend Percival about his New Theory of Vision, explaining that he has made some adjustments in light of King's (harsh) criticisms. He writes:
I met with some who supporting themselves on the authority of Archbishop of Dublin's [i.e. King] sermon concerning the prescience of God, denied there was any more wisdom, goodness or understanding in God than there were feet or hands, but that all are to be taken in a figurative sense; whereupon I consulted the sermon and to my surprise found his Grace asserting that strange doctrine. (Letter 12 [8], 35, [31f.])Footnote 21
While a consideration of the extent of King's influence on the early development of Berkeley's thought is beyond the scope of the present paper, the analysis of the sermon Berkeley mentions in the second part of the quote is important.
In his sermon on Divine Predestination and Fore-knowledge, consistent with the Freedom of Man's Will, King was concerned to defend the ‘Doctrine of Predestination’ (King, 1709, §1) against its apparent inconsistency with the ‘contingency of events’ (King, 1709, §6). While King agrees that humans could not consistently possess both infallible foreknowledge and free will, he thinks this inconsistency does not arise in the case of the divine. This is explained by the fundamentally different nature of God and the divine attributes compared to our own:
[I]t is in effect agreed on all hands, that the Nature of God, as it is in it self, is incomprehensible by human Understanding; and not only his Nature, but likewise his Powers and Faculties, and the ways and methods in which he exercises them, are so far beyond our reach, that we are utterly incapable of framing exact and adequate Notions of them. (King, 1709, §3)
According to King virtually everyone agrees that the divine nature and attributes are incomprehensible for humans and hence that we have no proper notion of them. It is in this context that the man born blind makes it first appearance in the sermon:
And if God's Foreknowledge and Predetermination were of the same nature with ours, the same Inconsistency would be justly infer'd. But I have already show'd that they are not of the same kind, and […] that they are quite of another nature, and that we have no proper Notion of them, any more than a Man born blind has of Sight and Colours; and therefore that we ought no more to pretend to determine what is consistent or not consistent with them, than a blind Man ought to determine, from what he hears or feels, to what Objects the Sense of Seeing reaches. (King, 1709, §7)
King uses the man born blind to illustrate his point that we have no proper notion of God and his attributes. This, in turn, is supposed to solve the apparent inconsistency. King's solution is, simply put, that it only seems like an inconsistency to us because we have only inadequate knowledge of the divine attributes (King, 1709, §30). As King sees it, just as a man born blind has no adequate notion of light and colours, we have no adequate notion of divine foreknowledge. Thus, we are in no better a position to give an accurate account of divine foreknowledge than a blind man is to give an accurate account of the nature of light and colours.
Admittedly, some of the details of King's description of a man born blind differ from Molyneux's original example (for example, King's blind man is not, at first, made to see). However, upon closer inspection it is clear these differences are superficial, and that King's use of the man born blind is more than a mere verbal coincidence. For example, King takes up what seems like a reversed version of the Molyneux problem when he denies the blind man could infer from tactile perceptions anything about what visual perceptions would be like. Moreover, King does draw a comparison with a blind man made to see several sections later. In §12, he explains that our knowledge of the divine attributes is, at first, equal to the blind man's knowledge of light and colours. However, our prospects with regard to God's attributes are more promising than a blind man's with regard to light and colours because we can hope to attain knowledge of the divine attributes in the next life (King, 1709, §12).
So far, we have shown that King employs Molyneux's example of the man born blind in relation to his account of our knowledge of divine attributes. This is not enough in itself to establish that King accepts the resemblance thesis. However, that this is indeed the case becomes clear if we consider his account of representation. Shortly after introducing the ‘man born blind’ in §7, King points out that
when we would help a Man to some Conception of any thing, that has not fallen within the reach of his Senses we do it by comparing it to something that already has, by offering him some Similitude, Resemblance or Analogy, to help his Conception. (King, 1709, §8)
In what follows, King illustrates this point by using the example of a map which, he argues, is a representation of the depicted country, in just the same way as our attributes are representations of the divine. As King points out, a map may represent the depicted country – a three-dimensional assembly of mountains, rivers and so on – despite being a two-dimensional sheet of paper. He explains that no one in their right mind would assume that countries are made of paper or that ‘China is no bigger than a Sheet of Paper, because the Map, that represents it [is]’ (ibid., §8). King argues that there need only be a ‘faint resemblance’ or a ‘little likeness”, as he later puts it, between two things in order for one to represent the other. Note, however, he thinks that such a resemblance must exist. In general, King thinks the resemblance between ‘Similitudes and Representations’ and what they represent
lies not in the Nature of them, but in some particular Effect or Circumstance that is in some measure common to both; we must acknowledge it very unreasonable to expect, that they should answer one another in all things. (King, 1709, §8)
For our purposes, it is not important to discern exactly what King means by ‘particular Effect or Circumstance’. What is important is that §8 clearly shows that King accepts the resemblance thesis. This should be clear from the fact that King advocates two claims. Firstly, when we wish to provide someone with a conception of that with which they are not acquainted, we use a resemblance or similitude of that thing. Secondly, a map's ability to represent is determined by the resemblance (however faint) it bears to a particular country.Footnote 22
What's more, we are now in a position to appreciate that King's treatment of the man born blind is grounded in his acceptance of the resemblance thesis. According to King, to a man born blind, ideas of light and colours do not fall ‘within the reach of his senses’. Thus, we can only attempt to provide a conception of light and colours to him by virtue of weak representations thereof. However, as King explains, these will inevitably be ‘imperfect Representations’ (King, 1709, §12). Like a map and the country it represents, there is no likeness in nature between sensations of light and colour and the notions of them had by a man born blind. The best that can be hoped for is similarity in ‘Effect and Circumstance’. What we have seen, then, is that King develops an account of knowledge of divine attributes grounded on the assumption that representation requires resemblance. Moreover, King uses the example of the blind man made to see to elucidate that account. In the next section, we demonstrate that despite challenging King's account, Berkeley likewise accepts the resemblance thesis and similarly draws on Molyneux's example.
4. The Resemblance Thesis in Berkeley
4.1 Divine attributes
In this section, we consider the influence of both Molyneux's man born blind and King's use of it in the context of knowledge of the divine attributes on Berkeley's own views. In this way, we trace the roots of the resemblance thesis in Berkeley's engagement with these two discussions which shaped the Irish intellectual milieu in Berkeley's time.
Although the previously quoted letter to Percival confirms Berkeley's early interest in the problem of divine attributes, Berkeley waited until the first publication of Alciphron (1732) to comment on it publicly and explicitly. In §§16–22 of the fourth dialogue Berkeley argues that the difference between divine and human attributes is one of degrees rather than one in nature. As one of his spokespersons Crito puts it:
But for your part, Alciphron, you have been fully convinced that God is a thinking intelligent being, in the same sense with other spirits, though not in the same imperfect manner or degree. (Alc. 4.22, p. 171)
Berkeley's use of the phrase ‘the same sense’ is important here, and indicates that, as he sees it, when we refer to either ourselves or God as ‘thinking intelligent being[s]’ we are doing so univocally which, in turn, is possible because the attributes are of the same nature. This is confirmed by several remarks in his earlier works. For example, in the Three Dialogues, he argues we can represent the nature of God to ourselves, via the immediate knowledge we have of our own minds (DHP, 3.231).Footnote 23 As Berkeley's spokesperson Philonous in the Three Dialogues explains:
my soul may be said to furnish me with an idea, that is, an image, or likeness of God, though indeed extremely inadequate. For all the notion I have of God, is obtained by reflecting on my own soul heightening its powers, and removing its imperfections. I have therefore, though not an inactive idea, yet in my self some sort of an active thinking image of the Deity. (DHP, 3.231–232).
Although Berkeley admits that it is to take the word idea ‘in a large sense’ when we understand our soul as an image of God (an infinite mind), the crucial point for our purposes is twofold. Firstly, this procedure only makes sense if human and divine attributes are of the same nature.Footnote 24 In other words, this only makes sense if Berkeley is using the term ‘mind’ or ‘spirit’ univocally as applied to God and ourselves.Footnote 25 Secondly, the procedure only makes sense if it is a case of representation in a strict sense – i.e. if it is not a case of signification. Yet, this does not entail that the resemblance relation is exactly the same as when ideas are concerned. Otherwise, Berkeley would not stress that we cannot ‘represent’ God or any other (finite) minds by ideas (DHP, 3.231).Footnote 26 However, it is equally evident that Berkeley thought that minds (finite or infinite) are alike. Berkeley explicitly connects the notions of ‘likeness’ and ‘image’ in relation to our knowledge of God.Footnote 27 In other words, the kinds of things as well as their likeness may be of a different kind when it comes to minds and ideas. Nonetheless, in each case representation requires resemblance and hence both are instances of representation in a strict sense.
In short, Berkeley, like King, accepts the resemblance thesis and brings it to bear on the problem of divine attributes. Berkeley agrees with King that human attributes can be considered as ‘representations’ of their divine counterparts precisely because there is some resemblance between them. However, while the two authors agree about the mechanism of representation, they fundamentally disagree about the nature of this resemblance. King does not distinguish between representation as instantiated in ideas and minds respectively the way Berkeley does. Moreover, he argues the ‘faint resemblance’ between God's attributes and our own is not grounded in a shared nature but in ‘circumstances and effects’ (King, 1709, §8). Berkeley, however, is happy to accept that God and finite spirits are similar in nature. He takes this to provide him with firmer foundation on which to argue that there is a relation of representation between the two. In fact, he claims to have rejected King's account for precisely this reason – viz., according to Berkeley, King's position does not allow us to prove the existence of (a wise and benevolent) God (Letter 12 [8], 36, [32]; Alc., 4.17–18, pp. 163–66).Footnote 28 Thus, Berkeley's approach constitutes a robust denial of King's account of our knowledge of the divine attributes.
It is therefore safe to assume when Berkeley refers to a ‘man born blind’ in the conclusion of §21, it is again more than a mere verbal coincidence:
This doctrine, therefore, of analogical perfections in God, or our knowing God by analogy, seems very much misunderstood and misapplied by those who would infer from thence that we cannot frame any direct or proper notion, though never so inadequate, of knowledge or wisdom, as they are in the Deity; or understand any more of them than one born blind can of light and colours. (Alc., 4.21, p. 171)
The conclusion of §21 certainly looks like a thinly veiled criticism of King. Contrary to King (1709, §12) Berkeley thinks we have a better notion of the divine attributes than someone born blind can have of light and colours. The disagreement between King and Berkeley, therefore, lies in the nature of, and extent to which, divine attributes resemble our own. However, neither contests that in order for us to represent to ourselves the divine attributes we need to be acquainted with something like them.
Berkeley's usage of the man born blind clearly illustrates how his solution differs from King's. Moreover, it is evidence of his engagement with his immediate intellectual milieu in Ireland in which, we have argued, the resemblance thesis was widely accepted. This engagement becomes more explicit in Berkeley's optical writings (the New Theory and Theory of Vision Vindicated and Explained) where the ‘man born blind’ resurfaces and plays a crucial role in his argument for the heterogeneity of the objects of vision and touch.
4.2 The heterogeneity thesis
Our final aim is to demonstrate that in discussions concerning Berkeley's acceptance of the resemblance thesis, we ought to broaden the scope beyond just the argument in §8 of the Principles. We have already demonstrated Berkeley's account of human knowledge of the divine attributes, developed in response to King's own account, is grounded upon the assumption that representation requires resemblance. In this section, our aim is to show that the resemblance thesis is also central to Berkeley's argument for the heterogeneity of visible and tangible objects in the New Theory of Vision (NTV) – despite the resembling things (i.e. ideas) being radically different than in the case of the divine attributes. Furthermore, Berkeley's frequent employment of Molyneux's example of a man born blind demonstrates that he is once again drawing on issues that have arisen in debates amongst Irish thinkers such as Molyneux and King in developing this argument.
Berkeley sets out with two aims in the New Theory, the second of which is to ‘consider the difference there is betwixt ideas of sight and touch, and whether there be any idea common to both senses’ (NTV, §1).Footnote 29 Ultimately, his answer is a negative one; there are no ideas common to both senses. Berkeley's claim, which he establishes over the course of the text, is that the constant connection that we perceive between certain visual experiences (e.g. seeing shapes with corners) and certain tangible experiences (e.g. the feeling of sharpness) is only a ‘habitual connexion that experience has made us to observe between them’ (NTV, §147). What is significant, for our current purposes, is that Berkeley frequently employs the man born blind example in order to illustrate the impossibility, as he sees it, of objects that are common to both senses.Footnote 30
In §132, Berkeley explains that his conclusion is confirmed by ‘the solution’, as he puts it, ‘of Mr. Molyneux's problem’. Berkeley is clearly convinced by Molyneux's and Locke's negative response to the question of whether a blind man made to see could differentiate between a cube and a sphere by sight. With that in mind, he explains,
if a square surface perceived by touch be of the same sort with a square surface perceived by sight, it is certain the blind man here mentioned might know a square surface as soon as he saw it. (NTV, §133)
For, if a tangible square and a visible square were of the ‘same sort’, presenting the (no longer) blind man with a cube would simply be ‘introducing into his mind by a new inlet an idea he has already been well acquainted with’. We are left with a dilemma, although not a very difficult one to get out of. Berkeley claims; either we allow that ‘visible extension and figures are specifically distinct from tangible extension and figures, or else that the solution of this problem given by those two thoughtful and ingenious men is wrong’ – the latter option is never seriously entertained. It is clear from this instance that Berkeley places considerable weight on the response to the problem provided by Molyneux and Locke.
In this way, Berkeley uses the ‘root metaphor’ of Early Modern Irish philosophy to confirm his conclusion that the object of sight and touch are entirely heterogeneous. The heterogeneity of tangible and visible objects, Berkeley goes on to explain, has important ramifications regarding the nature of representation. For example, he considers the potential objection that, since tangible squares are ‘liker’ (i.e. more similar) to visible squares in virtue of having ‘four angles and as many sides’, it follows that they are ‘of their own nature fitted to represent them, as being the same sort’ (NTV, §141). In response, Berkeley explains:
I answer, it must be acknowledged the visible square is fitter than the visible circle to represent the tangible square, but then it is not because it is liker or more of a species with it, but because the visible square contains in it several distinct parts, whereby to mark the several distinct corresponding parts of a tangible square, whereas the visible circle does not. (NTV, §142)
His point is that, in this case, the term ‘represent’ can only be used loosely in reference to an arbitrary relation; that is, one grounded on experience and custom. Only once it is agreed that a particular visible idea ‘represents’ a particular tangible idea in this way, can it be said of certain visual ideas that they are ‘fitter’ to stand for certain tangible ideas. However, this is not a case of representation in a strict sense, for it is not grounded in any relation of, what Berkeley later calls, ‘likeness or identity of nature’ (NTV, §147). The relation between the ideas of sight and touch is, then, not one of representation (in a strict sense) but of signification.Footnote 31
One reason for this is that by perceiving a certain visible shape (say, a visible square) one is not automatically imparted with a conception of the corresponding tangible shape (a tangible square). Berkeley goes on to confirm this by once again drawing on the example of a blind man. If a blind man were warned by a sighted person that he had come to ‘the brink of a precipice’, then he would justifiably be surprised by this apparent foreknowledge.Footnote 32 As Berkeley puts it,
He cannot conceive how it is possible for mortals to frame such predictions as these, which to him would seem as strange and unaccountable as prophesy does to others. (NTV, §148, our emphasis)
Berkeley's claim is the blind man cannot possibly conceive, or represent to himself, how this ability, of seeing what is ahead of oneself, could work. For he has neither prior experience of such an ability nor any ideas, conceptions, or experiences that are of the same nature or kind. The blind man's ability to represent the world to himself, Berkeley argues, is restricted by what he has experienced it to be like. Moreover, this is precisely the point he is making in Alc. 4.21 which can be understood as him saying that our notion of the divine may be inaccurate but not inexistent because light and colours are the proper objects of sight (PHK, §46; Alc., 4.10) and in virtue of lacking the ability to see and the heterogeneity of ideas, there is nothing the blind man could use to represent them to himself. In Berkeley's own words: ‘a blind man, when first made to see, […] would neither perceive nor imagine any resemblance or connexion between these visible objects [i.e. light and colours] and those perceived by feeling [i.e. distance]’ (TVV, §44).
Conclusion
We set out to explain Berkeley's acceptance of the resemblance thesis by placing his views concerning representation and resemblance in the context of debates that were happening in his immediate intellectual environment in Dublin and Ireland. Thus, we outlined the Irish roots of a principle which plays a crucial role in Berkeley's argument against representationalist epistemology in §8 of the Principles. We did so on the basis that if a contextual explanation is available, then before looking to the Cartesian or Aristotelian traditions, we ought to consider Berkeley's immediate intellectual context in Ireland. We then charted the progress of the resemblance thesis through Molyneux, King, and Berkeley's own active engagement in debates that shaped this Irish intellectual milieu. Molyneux's example of the blind man made to see, we argued, brought to the surface the notion that in order to conceive of an object one must already be familiar with something that resembles that object. We then suggested that the example of a ‘man born blind’ was employed by both King and Berkeley in their respective accounts of human knowledge of the divine attributes. While King and Berkeley disagree over the extent to which humans can be said to have knowledge of the divine attributes, neither contests the fact that this knowledge must be grounded on a relation of resemblance. Finally, we argued Berkeley's acceptance of the resemblance thesis plays a key role in his argument for the heterogeneity of the objects of sight and touch. It became evident that while things represented are of a different kind in each instance, in both cases, their representational relation requires resemblance, i.e. that these are cases of representation in a strict sense. In this way, we demonstrated that the resemblance thesis, while often implicit, can be identified in some of the most immediate and earliest influences on Berkeley's thought as well as important discussions within his work which, at first sight, may not seem connected. Thus, without ruling out alternative explanations, we conclude that Berkeley's acceptance of the resemblance thesis is rooted, primarily, in Irish thought.Footnote 33
