1. Introduction
The dedications inscribed above are both personal and directly relevant to the topic of this essay. My interest as a graduate student, over 40 years ago, in Aquinas’s theory of intentionality was encouraged (if with some initial scepticism) by my doctoral supervisor David Hamlyn, a wide-ranging philosopher who had succeeded Gilbert Ryle as editor of Mind—the oldest and then and still the leading British philosophy journal. Hamlyn translated and provided commentary on the De Anima for the Clarendon Aristotle series and wrote extensively on the subject of perception and on philosophical psychology and epistemology more generally. At Oxford, he had taken degrees in classics and philosophy, and in psychology and philosophy. His books devoted exclusively to perception began with The Psychology of Perception published in the same year (1957) and in the same Wittgenstein-oriented series (Essays in Philosophical Psychology) as Peter Geach’s Mental Acts, then continued with Sensation and Perception (1961) and concluded four decades later with Understanding Perception (1996).
Hamlyn described the last as ‘summing up my thinking’, and the most significant recurrent part of that thinking is that perception is conceptually informed, and that while the causal conditions of sense-cognition are non-conceptual ‘inputs’ to the sense faculties, they are not themselves cognitive components. In this, he was in agreement with Geach and Wilfrid Sellars in their critiques of the empiricist view that empirical knowledge is built up from primitive cognitive but pre-conceptual sensory givens.Footnote 1
In his De Anima commentary and in his works on perception, Hamlyn refers to Aquinas’s account, describing it as following the lines of Aristotle’s but differing in two significant respects. First, in focussing on the reception of sensible forms and the assimilation of senses and objects as pertaining to the sense faculties, rather than to the sense organs, thereby constituting sensuous apprehensions of those objects. Second, in providing an account of the process of transition from the actualisation of the sense-faculties in response to changes in the sense organs through to intellectual cognition of intelligible forms via the activity of two cognitive capacities: the cogitative power (vis cogitativa) and the active reason or agent intellect (intellectus agens). Hamlyn’s interest in the differences between Aristotle and Aquinas was related to his own philosophical concern with the nature of perception: recognising in Aquinas’s account, first, some awareness of the distinction between the physiological (sensation) and the epistemic (perceptual awareness), and, second, a recognition of the distinction between particular sensible, and general intellectual cognition.Footnote 2
It was while I was a student of Hamlyn that I published my first philosophy article ‘Aquinas on Sense-Perception’ (1983) responding to a paper by S.M. Cohen on the reception of sensible forms in which he criticised aspects of Hamlyn’s account of Aquinas’s view as expounded in Sensation and Perception.Footnote 3 My concern was in part with Thomas’s theory of how singular reference in thought is achieved given that on his theory concepts are universal in content. His solution is that by turning to stored sensory forms derived from experience (conversio ad phantasmata), a particular individuating dimension is added. But the question then is whether, or in what sense, the reception and ‘storage’ of sensible species is physical. Resolving this question turns in part on the interpretation of such passages as this:
In the lower terrestrial natures there are two degrees of immateriality. There is the perfect immateriality of intelligible being; for in the intellect things exist not only without matter, but even without their individuating material conditions, and also apart from any material organ. Then there is the half-way state of sensible being. For as things exist in sensation they are free indeed from matter, but are not without their individuating material conditions, nor apart from a bodily organ. For sensation is of objects in the particular, but intellection of objects universally.Footnote 4
In the intervening period, the subject of Thomas on sense-cognition has been much discussed, including the business of the role of sensible species and the nature of their reception in the senses; but 40 years ago, there was very little published. In the International Bibliography of writings on Aquinas between 1977 and 1990, the term ‘perception’ does not occur in the index of items in English, and ‘sensation’ only has one item under it. In the index of foreign terms, there is only one entry: ‘percepión’ linking to a 1978 Spanish edition of Cornelio Fabro’s 1941 book Percezione e Pensiero.Footnote 5 Since then, by far the most substantial scholarly and philosophical work on the subject is the late Tony Lisska’s Aquinas’s Theory of Perception: An analytic Reconstruction (2016).Footnote 6
I first met Tony at one of Ralph McInerny’s summer Thomistic Institutes at Notre Dame in the mid 1990s and was immediately taken with his enthusiasm, depth of reading in Aquinas and intense desire to connect Thomism with contemporary analytic philosophy. This ambition had already born fruit with his book Aquinas’s Theory of Natural Law: An Analytic Reconstruction (1996) preparatory work for which had been done while on a sabbatical at Oxford where he was encouraged by Anthony Kenny and Brian Davies OP.
In 1998, I was a commentator on an ACPA conference paper ‘Why the Cogitative Power?’ by Leo White who had recently got his PhD from Catholic University of America for a thesis on The Experience of Individual Objects in Aquinas. I said to Tony that White’s essay, in which he argues that for Thomas the cogitative acts as a bridge between the sensory grasp of particulars and the intellectual grasp of universals, had led me to see that the vis cogitativa was an important, if obscure, but generally neglected aspect of Aquinas’s views on cognition, and specifically on the relationship between sense-perception and thought.Footnote 7
It was then standard to explain the cogitative power as being the counterpart in human beings of the estimative power in animals (vis estimativa). That is true, of course, but what seemed much more important, and then largely undiscussed, was its role in Thomas’s account of concept-formation and the exercise of concepts in perception. This had Tony excited, and the following year he gave a paper to the International St Thomas Aquinas Society on the ‘The Vis Cogitativa and the Perception of Individuals’ adding to his resolve to produce a study of Aquinas’s theory of perception in which the cogitative power would be shown to be a critical component, a matter to which I will return.
Tony was formed by the Dominicans: educated at Aquinas College High School in Columbus Ohio, and at Providence College, Rhode Island, and for a period was a Dominican scholastic in training. Like Hamlyn, he was a dedicated teacher and he will be long-celebrated for his excellence as a mentor of undergraduates for over half a century at Denison University, and for his contribution to the exploration of Aquinas’s accounts of natural law and of perception. His and David Hamlyn’s intellectual worlds were quite distinct, but it was my very good fortune to have inhabited both, and I am grateful for it.
2. The philosophy of perception
Historical and exegetical scholarship apart, if one were to look today for a comprehensive account or theory of human visual perception,Footnote 8 it would be in a work drawing on several fields of study including geometric, physical and ecological optics, physiology, neuropsychology, and cognitive perspective and colour science. This raises a couple of salient questions. First, why, given these increasingly well-developed and progressively integrated empirical disciplines, should there also be a place for philosophy in such a comprehensive account? Second, how does Aquinas’s theory of perception (as set out in Summa Theolgiae 1a q.78, 3 & 4. De Veritate q,10, and in his Commentaries on Aristotle’s De Anima II, and De Sensu) relate to modern and contemporary scientific and philosophical theories?
There are three related answers to the first question. The most familiar is that what needs to be understood is the general nature of perception in its various modes and in its relation to other cognitive functions such as imagination and intellection, and the difference between being impacted by aspects of one’s immediate environment and seeing things in it; and the route to understanding these issues is not by further empirical investigation but by conceptual analysis. The second answer extends the first, by observing that each of these empirical disciplines gives rise to highly abstract general questions that transcend each and all of the sciences but that fall within the sphere of philosophy: questions about the nature of causality, and the relation between physical processes and mental states. The third type of answer is that what is needed for understanding perception is an account of the conditions of its possibility. How can it be that standing some miles away across a Scottish glen and on the other side of a forest I am able to see a Castle through gaps between the trees, and see that it is massively larger than my hand even though if I raise my hand in front of me it occupies a much larger part of my visual field? Of course, empirical facts of the sort studied by the sciences of perception are relevant to answering aspects of this question, but there is something else: the matter of cognition—knowledge of reality.
Perception is a way of coming to know things and is subject to standards of correctness and accuracy. Thus, it falls within what Sellars termed ‘the space of reasons’ writing that ‘in characterizing an episode or a state as that of knowing, we are not giving an empirical description of that episode or state; we are placing it in the logical space of reasons, of justifying and being able to justify what one says’.Footnote 9 This is an insight that John McDowell has made much of in his own work in epistemology and philosophy of mind.Footnote 10 Perception is also connected to how we conceive of things and what we believe about them, both contributing to conception and belief and being influenced by them. How, then, can empirical facts about causal processes involving ambient and reflected light, the stimulation of the retina, the optical tracts and the brain explain the way things look to a subject in a sense of ‘look’ that involves conscious experience and perceptual judgement? What exactly are the immediate objects of visual experience? Wherein lie the standards of correctness for perceptual judgements, and do they differ as between different objects of perception such as colour and shape? If so, why? These are all philosophical questions.
In saying this, however, I do not wish to suggest, as was commonly held in the heyday of conceptual analysis, that philosophical and empirical issues are wholly separate. One way in which such a distinction used to be drawn was by saying that philosophical questions concern necessities arising from the relations between the meanings of terms; whereas scientific questions concern contingencies relating to empirical objects and causal relations between them. But such a dualism is untenable, since concepts about the world express our understanding of it as formed through (and by reflection upon) our experience of things and the relations between them, many of which are non-contingent.
Better, then, to say that semantically the contingent and the (non-logical) necessary stand along a spectrum in which they are mixed to various degrees, rather as hues of green become progressively yellow(ish) and blue(ish) as they proceed in opposite directions towards distinct colours. The purported absolute distinction between the necessary and the contingent was standardly cast in terms of a contrast between the analytic (true/false purely in virtue of meaning) and the synthetic (true/false in virtue of worldly facts). This doctrine was famously demolished by Quine in the 1950s,Footnote 11 and efforts to rebuild it have never succeeded in showing more than that there is perhaps a distinction of relative degree between the empirical and the conceptual. This suggests a different approach, which is to say that the appearance of linguistic necessity or non-contingency is not due to concepts or terms in isolation from reality but to them as representing features of it. This is precisely what Aquinas held in saying that it belongs to the ratio (intelligible content) of certain concepts that what they represent is something necessary or, and this is the case in the case of perception, that what it represents is something that obtains for the most part ‘ut in pluribus’, where this is not a statistical but a categorial matter.Footnote 12 For contemporary readers, the point is perhaps more perspicuously presented by saying that it is part of the concept of perception (and of the conditions of its application) that its objects are external realities. Of course, one may mis-perceive in various ways, but the notion of misperception is parasitic upon that of veridical perception, and the possibility of misperception is investigated and confirmed by veracious observation. More generally, and more fundamentally, the terms in which perceptual contents are describable, even in the case of hallucinations, belong to the vocabulary of publicly observable objects and features. In sum, the general reliability of perception as a mode of objective knowledge is not a contingent matter but a condition of the possibility of understanding and explaining deviant cases.
3. Explanatory not justificatory
How in light of this should we think of what Aquinas has to say about perception? Is it scientific or philosophical in nature or a combination of both, or something else? Part of the difficulty in answering this question is that there is no clear alignment between his notions and ours. For one thing, he would not distinguish between natural and speculative philosophy in the way in which modern and recent thinkers have done. A more precise and directly relevant example, however, is the idea of experience and its relation to cognition. In modern philosophy, dating from Descartes and Locke and continuing to the present day, a distinction is drawn, or presupposed, between subjective conscious awareness and objective perceptual knowledge, such that no amount of the former guarantees anything of the latter: hence the issue of external world scepticism. Many efforts to establish a bridge between awareness and knowledge have mostly assumed a lowest common denominator notion of experience, to the extent that it seems difficult to conceive of sense-perception as anything other than a subjective psychological state (conscious awareness) plus some further non-psychological condition such as the former being caused in a certain away by external realities.
If we look to Aquinas, however, it seems that he does not have this philosophical concept of experience. The apparent candidates for equivalents are experimentum, experientia, and cognitio experimentalis, but in the contexts in which he uses these terms, they refer to cognitive rather than subjective states: observation, non-discursive sensory knowledge, and experiential knowledge in contrast to knowledge of first principles and theoretical reasoning.Footnote 13 By way of example, in Summa Theologiae 2a 2ae, q. 95 a5, he writes that ‘human science originates from experiments [ex experimentis], according to the Philosopher (Metaphysics. i, 1). Now it has been discovered through many experiments [multa experimenta] that the observation of the stars is a means whereby some future events may be known beforehand’. Two questions later (q. 97, a2, ad 2), he remarks that ‘There is a twofold knowledge of God’s goodness or will. One is speculative … The other knowledge is affective or experimental [cognitio … affectiva seu experimentalis] and thereby a man experiences in himself the taste of God’s sweetness’Footnote 14 What is at issue in the first case is observation understood as the perception of various objective realities, and in the second interior awareness of the divine. Again, in De Malo he writes that ‘Experience in the strict sense belongs to the senses [experientia proprie ad sensum pertinent]. For although the intellect knows both separate forms, as the Platonists held and material substances, the intellect knows these things by their general nature, not as they exist here and now, which is to have experience [quod est proprie experiri]’. And a few articles later states that ‘Experiential knowledge derives from sense perception [experientia procedit ex sensu] as the senses know things that are present [to them]’.Footnote 15 ‘Experience’ in these uses corresponds more or less to the everyday sense of observation or encounter, not to the modern philosophical notion of a subjective phenomenological state characterisable without reference to any independent reality.
Not having the latter notion, Aquinas is not troubled by the question of the relation between experience and knowledge in the form that would render his account vulnerable to the spectre of scepticism. This also explains the difference between the kind of account of perception that he is engaged in providing, which is essentially descriptive-cum-explanatory, and that which Descartes, Locke and their successors through to the present day are concerned with, which is epistemological in the modern sense of seeking to justify the claims of sense experience to attain to the status of authentic knowledge. If one were to look for a recent philosophical counterpart of Aquinas’s conception of the task of explaining perception and its place in the general framework of knowledge it would, ironically, be Quine’s notion of naturalised epistemology. In Quine’s words:
Epistemology, or something like it, simply falls into place as a chapter of psychology, and hence of natural science. It studies a natural phenomenon, viz., a physical human subject. This human subject is accorded a certain experimentally controlled input – certain patterns of irradiation in assorted frequencies, for instance – and in the fullness of time the subject delivers as output a description of the three-dimensional external world and its history.Footnote 16
The verbal and phonic proximity of ‘D’Aquino’ and ‘Quine’ apart, the irony is that Quine was quasi-behaviourist, physicalist and atheistic, yet both he and Thomas conceived of knowledge in ways that were free of sceptical doubts. For Quine’s ‘patterns of irradiation in assorted frequencies’ substitute Thomas’s ‘sensible forms passing from objects via an intervening medium and affecting the [external] sense organs’. What happens next in Aquinas’s account involves the internal senses (interiores sensus): the common sense, the imagination, the cogitative, and the memorative powers. His description of these as powers is more accurate than ‘senses’ (which he also uses) since they are capacities to work on sensory inputs, not themselves sensory sources. For present purposes, the details of their operation do not matter, but what is relevant is that the cogitative power plays a role in respect of sense cognition analogous to the played by the active intellect in relation to intellectual cognition. Aquinas writes:
The action of the cogitative power consists in comparing, adding and dividing [sensory contents] … Wherefore it is also called the ‘particular reason’ … for it compares individual intentions [sensory representations] just as the intellectual reason compares universal intentions [conceptual representations]. … The cogitative and memorative powers in man owe their excellence not to that which is proper to the sensitive part; but to a certain affinity and proximity to the universal reason which, so to speak, overflows into them.Footnote 17
Two points are of importance. First, that the transition in Aquinas’s account from sensory input to, in Quine’s phrase, ‘a description of the three-dimensional external world’ depends critically on the cogitative power’s ability to compare and synthesise the products of the various sense faculties, and thereby produce complex perceptual representations upon which the active intellect can work to produce abstract concepts. Second, intellectual conception somehow flows back into the perceptual level.
4. Empiricalism not empiricism
Aquinas is of the view that all human knowledge derives from experience. Given, however, that what exists at the sensory level is particular and individuated by the materiality of the sense-organs, and that what exists at the intellectual level is universal and entirely apart from materiality, he has to give a theory of the conditions and processes involved in the movement from sensory awareness to intellectual apprehension. He believes that imagination, memory and the common or unifying sense are insufficient to provide the active intellect with something suitable for forming a universal content apt to be the formal principle of intellectual acts. Therefore, he postulates a cognitive power lying someway between sensory awareness and intellectual judgement to affect the transition between them: cogitation. If I am to judge that the object before me is a human being, then an individual has to be identified and brought under a universal concept: I have in some way to be receptive to the input to the sense while also being active in discerning what kind of thing it is.
Here, an anti-abstractionist objection may suggests itself: singling out an individual as a human being presupposes possession of the concept human and, hence, cannot be part of the explanation of the formation of the concept.Footnote 18 In reply, it might be said that the cogitative power does not single out an individual as-a-human-being, but rather that it singles something out that is a human being. That it does the latter may be true but that cannot be an accident or else the groundwork for the formation of the concept human being will not have been done. The answer lies, I think, in viewing the attainment produced by the cogitative power as proto-conceptual. Consider the capacity of animals and pre-conceptual infants to recognise two-dimensional objects or outlines on the basis of their visible shape, the look of such things as they appear when oriented in a plane parallel to the viewer. Let me term this a ‘perceptual attainment’. Like a conceptual capacity, it confers the ability to identify and individuate things of a given kind, but, unlike the former, it is confined to sensible appearances and to favourable classes of those; so, the kind in question is a visual one. A wire triangle may be oriented to appear as a line and the perceptual attainment may then be insufficient for its recognition. By contrast, the concept ‘triangle’ makes no reference to how triangles look but to what a triangle (the abstract geometrical form) is. The shift from the cogitative to the intellectual is from sensible to intelligible cognition, from acts defined in terms of appearances to ones specified in terms of conceptual contents. How that is effected is by the mind making concepts that interpret, manipulate and transcend sensible appearances. In the question following that in which he introduces the cogitative power and speaks of its affinity and proximity to universal reason (conceptual thought) Thomas writes of the process of concept-formation using the analogy of illumination:
There are two opinions as to the effect of light. For some say that light is required for sight, in order to make colours actually visible. And according to this the active intellect is required for understanding, in like manner and for the same reason as light is required for seeing. But in the opinion of others, light is required for sight; not for the colours to become actually visible; but in order that the medium may become actually luminous, as the Commentator says on De Anima II [7 418a26]. And according to this, Aristotle’s comparison of the active intellect to light is verified in this, that as it is required for understanding, so is light required for seeing; but not for the same reason.Footnote 19
This is a somewhat obscure passage in part because Aquinas is negotiating his way through a thicket of existing ideas including a) conflicting theories about the nature of light, b) a tendency of some then-contemporary Augustinians (such as Bonaventure) to relate the process of concept acquisition to the idea of divine illumination, c) differing views of Aristotle’s account of the relationship of light to colours (the ‘Commentator’ referred to is Averroes) and d) the implications of c) for the interpretation of analogy of concept-formation to illumination proposed by Aristotle at De Anima III, 5 430a12. Thomas’s claim is that the analogy is only valid if we understand the illumination effected by the active intellect in its orientation toward sensory representations as not revealing something already present but as producing something new: the intellect creates intelligible forms rather than disclosing them. But what is created by the mind subsumes and makes sense of what was previously only recognised in respect of sensible characteristics. And once attained, the conceptual may flow back into the sensory to achieve conceptually informed perception, as in seeing a human being as a human being.
The issue of concept-formation is crucial to the question of whether it is correct to say, as many do, that Aquinas is an empiricist. Hamlyn is unequivocal on the matter: ‘while it is doubtful whether Aristotle can be called without qualification, an empiricist, there is no doubt that Aquinas was one. He was anxious, that is, to show that all the materials for knowledge … are to be derived from sense experience’.Footnote 20 Lisska, referring at several points to Hamlyn’s interpretation, rejects the empiricist label and claims that Aquinas recognises a distinction that Hamlyn insists upon in his own anti-empiricist view, namely, that between sensation and perception: ‘It is with the vis cogitativa that Aquinas goes beyond the analysis of perception proposed by Berkeley, Hume, and most empiricists in modern and contemporary philosophy. … [he accepts] the distinction between sensation and perception [and in] addition he would argue that a category difference exists between these two types of sense knowledge’.Footnote 21
Here and throughout his book Lisska attributes a great deal to the cogitative power, seemingly crediting it with providing kind-recognition antecedent to the work of the agent intellect: not just seeing proper and common sensibles (colour shape, etc.) but observing, as ‘incidental objects’ of sensory awareness, instances of substantial kinds, e.g., seeing a man.Footnote 22 I think this oversteps, in the wrong direction, the line between cogitation and intellection and am inclined to attribute such substance perception to what Aquinas says about universal reason overflowing into the cogitative power. It is indisputable, however, that Aquinas maintains that while human knowledge derives from sense experience it also transcends it, not just in dealing with non-empirical matters but even in judgements relating to the empirical where these apply intellectual conceptions. In addressing the question, ‘Whether Intellectual knowledge is derived from sensible things’, he writes:
[O]n the part of the phantasms [sense traces of experience], intellectual knowledge is caused by the senses. But since the phantasms cannot of themselves affect the passive intellect, and require to be made actually intelligible by the active intellect, it cannot be said that sensible knowledge is the total and perfect cause of intellectual knowledge, but rather that it is in a way the material cause. … Sensitive knowledge is not the entire cause of intellectual knowledge. And therefore it is not strange that intellectual knowledge should extend further than sensitive knowledge.Footnote 23
In summary, then it might be most apt to say that Aquinas is an empiricalist (knowledge derives from experience) but not an empiricist (all knowledge is reducible to experience).
5. Realism
Throughout the twentieth century, and continuing today, it has been common for admirers of Aquinas to refer approvingly to his ‘philosophical realism’. The context for this is the belief that modern philosophy, beginning with Descartes and Locke, and extending into recent times, is anti-realist: either sceptical, allowing that there may be an external world but denying that we know anything other than our own mental states; or constructivist, asserting that there is no such thing as a wholly mind-independent reality, that the ‘world’ is in some sense, and in part or in whole, an artifact of human thought and experience.
The celebration of Aquinas as a champion of realism is, however, surprisingly recent. In his popular book on Aquinas published in 1933, G.K. Chesterton gives voice to the idea of Thomas as a defender of common sense realism.
The fact that Thomism is the philosophy of common sense is itself a matter of common sense. Yet it wants a word of explanation, because we have for so very long taken such matters in a very uncommon sense. … Since the modern world began in the sixteenth century, nobody’s system of philosophy has really corresponded to everybody’s sense of reality; to what, if left to themselves, common men would call common sense. … [M]y only object is to show that the Thomist philosophy is nearer than most philosophies to the mind of the man in the street.Footnote 24
A few pages later he clarifies the point
Needless to say I am not so silly as to suggest that all the writings of St Thomas are simple and straightforward, in the sense of being easy to understand. There are passages I do not in the least understand myself; there are passages that puzzle much more learned and logical philosophers that I am; there are passages about which the greatest Thomists still differ and dispute. … The only point I am stressing here is that Aquinas is almost always on the side of simplicity, and supports the ordinary man’s acceptance of ordinary truisms. For instance, one of the most obscure passages, in my very inadequate judgement, is that in which he explains how the mind is certain of an external object and not merely of an impression of that object; and yet apparently reaches it through a concept, though not merely through an impression. But the only point here is that he does explain that the mind is certain of an external object.Footnote 25
Chesterton’s publisher and biographer, Maisie Ward, reports how when they learned that he was writing a book on St Thomas her husband ‘felt a faint quiver of apprehension. Was Chesterton for once undertaking a task beyond his knowledge?’ She continues
Such masses of research had recently been done on St Thomas by experts of such high standing and he could not possibly have read it all. [His secretary Dorothy Collins later reported that] He began by rapidly dictating about half the book. So far he had consulted no authorities but at this stage he said to her: “I want you to go to London and get me some books”. “What books?” asked Dorothy. “I don’t know,” said GK. She wrote therefore to Father O’Connor [the priest who inspired the figure of Father Brown, Chesterton’s most famous literary creation] and from him got a list of classic and more recent books on St Thomas. Footnote 26
Most volumes then available in English were collections of essays celebrating different aspects of Aquinas’s character and achievements, though there were more substantial works, principally Martin De Wulf’s Scholasticism Old and New (1907), Francesco Olgiati’s The Key to the Study of St Thomas (1925), Martin, Grabmann’s Thomas Aquinas (1928), and Martin D’Arcy’s Thomas Aquinas (1930). D’Arcy acknowledges those named as ones on whom he relied but explains that ‘the scope of this volume will however be found, I think, to differ from them. … I have therefore tried to present his philosophy in its unity in the light of its fundamental principles’.Footnote 27 Part II begins with a Chapter 4 entitled ‘The First Principles of Knowledge’. There D’Arcy writes:
Amongst the truths [which it is impossible to doubt or suspend judgement about while searching for some more ultimate truth] is that we know reality. Aquinas belongs, therefore, to what is called the dogmatist tradition in philosophy, and he holds in one sense that the so called problem of knowledge is a false problem. Knowledge cannot be justified by anything else save knowledge and in its act it possesses its own justification. Footnote 28
Comparing the two books, I suspect that Chesterton may have skimmed this chapter of and in that way got something of what Etienne Gilson had discussed in the first of his books listed in D’Arcy’s bibliography, namely, Le Thomisme. In 1929, an authorized English translation of the revised 3rd edition had been published under the title The Philosophy of St Thomas Aquinas. In it Gilson addresses Aquinas’s account of sense-knowledge, intellection and truth, but only as part of a broader treatment of nature, and the index has no entry for ‘realism’. In the same year, however, he began a series of articles on Thomistic realism, taking issue with some contemporary neo-Thomists, generally described as ‘critical realists’. These essays were later published in two books: Le réalisme méthodique (1936) and Réalisme thomiste et critique de la connaissance (1939).
Most salient in the present context is the second of these, subsequently published in English as Thomistic Realism and the Critique of Knowledge. The title is deliberately challenging, but more striking is that the first chapter in titled, ‘Realism and Common Sense’, and it begins in a markedly Chestertonian style. Gilson writes that ‘After passing twenty centuries as the very model of those self-evident facts that only a madman would ever dream of doubting, the existence of the external world finally received its metaphysical demonstration from Descartes’.Footnote 29 Immediately he explains the influence of the Cartesian critical turn, beginning not with things but with sensations and ideas, and then refers approvingly to the founder of the Scottish school of common sense philosophy:
When Thomas Reid recounted this remarkable story he was one of the first to discern its meaning, and it was his intent to escape the magic circle in which philosophers since Descartes had been trapped, mesmerized by the cogito and idealism without ever managing to get out. It was in large measure his resolute rejection of the Cartesian approach that led Reid to elaborate his doctrine of “common sense”.Footnote 30
I will return to Reid briefly in a moment, but note another important manifestation of the growing celebration of Aquinas as an epistemological realist was Jacques Maritain’s Les degrés du savoir (1932) published in English 5 years later.Footnote 31 This is a more systematic and detailed account of the theory of cognition to be found in Aquinas and in his scholastic commentators, principally John of St Thomas. What I want to suggest, however, and I have never seen this proposed or considered, is that the style and the framing of Gilson’s presentation of Thomistic realism may owe something to his reading of Chesterton’s book—in particular the emphasis on common sense realism vs the madness of philosophical denials of it. He later wrote ‘I consider it as being without possible comparison the best book ever written on St Thomas’.Footnote 32
If this is right, it may be that Gilson’s highly influential rejection of quasi-Cartesian and quasi-Kantian formulations of Thomistic cognitive theory is indebted, if only in its confident style and popular appeal, to none other than the author of the Fr Brown detective stories. This raises a question, however, of why it is that neither Gilson nor others prior to the 1920s and 30s made much of the idea that Aquinas was especially distinguished in being a ‘realist’ and even a ‘common sense realist’. Part of the answer is that the terminology was introduced to give a name to the opposite of views they were particularly concerned to repudiate, namely, forms of subjectivism and relativism that they found particularly threatening to natural theology and natural law ethics and that they suspected of entering Catholic thought through Marechal and the ‘transcendental Thomist’ movement. Others familiar with Anglo-American philosophy might have noted similar tendencies in British phenomenalism, and in the attack on ‘naieve realism’ developed by Roy Wood Sellars and other north American philosophers who made a case for experience being interpretative rather than simply presentational. It may have seemed timely, therefore, to invoke Aquinas on the side of real realism. Thanks to Aeterni Patris and its consequences, Thomas was for Catholic thinkers the philosopher/theologian of first choice, especially those involved in apologetic debates, as D’Arcy and Chesterton were. It would be appealing, therefore, to view him as avowing and giving centrality to an unambiguous and uncompromising common sense realist.
6. Was Aquinas really a realist?
I now turn to the question of whether this representation of him is accurate: was Aquinas really that kind of realist? Or was he indeed a realist at all? To answer these questions, we first need to get clearer about what realism amounts to. In general, however, talk of ‘isms’ is as misleading in relation to philosophy as it is with regard to art or to any other field of refined and discriminating practical or speculative production. Moreover, realism and anti-realism only make sense, if they do, in relation to some area or subject matter.
The celebration of Aquinas to be found in the writings of Chesterton, Gilson, and Maritain is primarily in relation to his epistemological position. Realism of this sort, which is advanced by Thomas Reid on the basis of an analysis of the Cartesian and Lockean theories of ideas, is opposed to anti-realist representationalism.Footnote 33 Advocates of the latter positions claim that the immediate, and perhaps the only objects of cognition are subjective states: impressions, sense-data, ideas, or whatever. Against this realists hold that whether or not there is that kind of subjective state cognition, there is also direct knowledge of things themselves. Reid writes of how ‘all philosophers, from Plato to Mr Hume, agree in this, that we do not perceive external objects immediately and that the immediate object of perception must be some image present to the mind’.Footnote 34And he opposes to this a theory of ideas as mental acts in which thoughts and perceptions are conceptually informed operations:
Conceiving, as well as planning and deciding, are what the schoolmen called ‘immanent’ acts of the mind, which produce nothing beyond themselves … Let this, therefore, be always remembered, that what is commonly called the image of a thing in the mind, is no more than the act or operation of the mind in conceiving it.Footnote 35
It is unfortunate, in view of his mention of the scholastics, that Reid does not exempt Aquinas from the charge of representationalism, since there can be little doubt that what he writes would place him on the realist side of the ‘ideas debate’:
The intelligible species is to the intellect what the sensible species is to the sense. But the sensible form is not what is perceived but rather that by which sense perceives. Therefore, the intelligible species is not what is actually thought of [and more than the sensible species is what is actually seen] but is that by which the intellect thinks.Footnote 36
What, then, of other kinds of realism and anti-realism? One that has been prominent in recent decades in analytical philosophy is semantic anti-realism. This is related to, though usually makes a point of distinguishing itself from, verificationism: the view that the meaning of a sentence is given by, and cannot transcend the possibility of its verification. On this account the sentence “The University of St Andrews was founded in 1413” is only meaningful, and therefore only apt to be evaluated as true or false, if it is possible to relate its content to some means of establishing conclusively whether it is in fact true or false. Now establishing the truth (or falsity) of a sentence will involve either proving it to be such on the basis of deducing this by logically valid inference from other sentences known to be true, or else confirming it by experience, or perhaps in a third way on the basis of testimony, though for the verificationist this will not be an independent method of determination. In the absence of any of these ways of showing a sentence to be true it lacks meaning.
Post-verificationist semantic anti-realism has a more relaxed attitude to proof and evidence, for one thing extending these to ‘in principle’ confirmation. Still the claim is that one cannot give meaning to sentences the conditions of whose truth could not in principle be known. While Aquinas does not consider this sort of argument, he has the resources to respond to it by pointing out that what is necessary for giving meaning is not being able to determine the truth of a sentence but to understand its content through a grasp and use of the concepts that feature in it, and that this may done through a variety of methods: abstractive induction, analogy, imagination, the combination of relative expressions, and so on. For Aquinas, there are connections between meaning and truth but they are not of any single or simple variety and we acquire and manifest our understanding in a large number of different ways, often by relating one claim to others and relating these to action.
Prior to its use in relation to knowledge and to meaning, the term ‘realism’ was introduced to describe positions on the question whether there is generality outwith the mind: are there in addition to individual things also general natures and characteristics? Are there as well as this black cat, blackness and catness? Debates about these sorts of questions as they were engaged in by medieval philosophers, including Aquinas, involve two broad positions: realism and anti-realism, further divided into two subdivisions. In this scheme there are radical and moderate realisms: the former holding that general natures exist prior to and independent of particular instances; and the latter, that while general natures exist they only do so in and through their instances. Against these are two opposing positions Radical anti-realism holds that the only generality there is that associated with the attaching of a common name to a group of individuals (hence ‘nominalism’). The moderate position (conceptualism) maintains that while there is no generality in nature nor is the meaning of general terms such as ‘black’ or ‘cat’ reducible to, or identifiable with the use of common terms. Rather, general natures or universals do indeed exist not in extra-mental reality but in the mind where they either pre-exist as innate universal concepts or come to exist a posteriori through intellectual formation. Where does Aquinas stand in relation to this debate? He writes regarding the example of the nature of essence ‘man’
This nature has a twofold being: one in individual things and the other in the soul one in individuals. … In individuals, moreover, the nature has a multiple existence corresponding to the diversity of individuals; but none of these beings belongs to the nature from the first point of view, that is to say when it is considered absolutely. … It remains, then that human nature happens to have the character of a species only through the being it has in the intellect.Footnote 37
His position is that natures are plural in things but single as formed and expressed in abstract concepts. In the world there are only individual human beings each with its own individual nature, but considering these ‘many’ we may form a conception in which these natures are considered apart from their material individuating conditions and this conception is of a general nature. People and their natures are many, humanity as an abstracted form is one. Moreover, in then deploying this concept to refer to individuals we do not err in subsuming them within a common class, and in doing so we are not merely inventing and attaching a name to an arbitrary collection of particulars, for the concept has to answer to the natural conformity of one cat nature to another.
Understood in this way we might say that Aquinas is less of a realist about common natures than Plato or Aristotle but more of a realist than say Ockham; yet this way of speaking is also misleading since it would be better to say that they differ in where they locate the reality of general natures. This brings me to two related but somewhat different debates to which ‘realism’ and anti-realism’ have been attached. So far as concerns metaphysics or ontology as these are classically understood, there are two questions one might ask about members of some kind K, or about the kind itself: 1) do Ks exist? or does K exist? and 2) how do Ks? or K exist? A ‘denier’ holds that Ks or K do not exist at all, whereas an ‘avower’ maintains that they do. But there can be a further difference in so far as some avowers may hold that Ks or K exist extra-mentally, whereas other avowers may hold that while they certainly exist, their existence is mind- or subject-dependent. We might then say that Aquinas is an avower as regards general natures but holds them to be mind-dependent. Is he then a realist or a non- or anti-realist? How to answer this depends less on what Aquinas actually says than where one wishes to place him in relation to other positions. Here again the language of isms’ shows its frailty and distorting tendency, since it makes it sound as if a position is an absolute one whereas it may be better seem as comparative or contrastive—certainly best seen in relation to a particular question rather than expressive of a general disposition.Footnote 38
Finally, I come to the second of the related debates and to another two questions: 1) What is known? That is, what are the objects of knowledge? and 2) How are they known? Like Reid, Aquinas holds that cognitive acts engage mind-independent objects via the mediation of sensory and ideational structures: ‘percepts’ and ‘concepts’. The shared belief of the two Thomases is that, save under conditions of reflective attention to the activity of sensory or intellectual cognition, these structures are not themselves the objects of cognition but are the things that shape and direct the mind towards the world. So far so anti-representationalist or anti-anti-realist; but what about the second question: how are things known? More precisely, are they known as they are in themselves or as we represent them?
Aquinas is apt to distinguish between a nature as it is according to the thing itself and according to the manner in which it is in the knower. The usually intended meaning of this is to distinguish between extra-mental and mental existence, as in the distinction between the way the catness of the cat is in the cat and the way in which it (or its universalized form) is in a thinker thinking of cats. But we should now recognize a further possibility. Recall that for the nominalist there are only individuals and names by which a plurality of the former are co-identified. Aquinas’s position is that the nominalist misses out another part of mind-independent reality, viz. the natures and characteristics of individual things. If there are two men, Peter and Paul, then there are also two individual natures: the humanity-of-Peter and that of Paul, and there are also various individual characteristics: the shape, size, weight etc. of Peter, and those of Paul. For Aquinas, then, general natures and characteristics exist but only in a mind-dependent way; and what exist mind-independently are all individuals: substances and property instances.
Traditional Thomists will allow this but not think it has further implications for debates about realism; I am not sure that this is so. Thomas’s realist champions will note that he says what exists in the mind in consequence of the operation of the active intellectual power is the general essence. Yet, he also tells us that we really know very little about the intrinsic nature of things. What, then, might constitute the content of the concept of a general kind? He writes ‘because substantial forms, which in themselves are unknown to us, are known by their accident; nothing prevents us from sometimes substituting accidents for substantial differences.’Footnote 39
Linguistic psychologists and philosophers of language have become interested in general grammatical forms of a particular sort with which Aquinas was very familiar.Footnote 40 Generalisations regarding a kind, K, having a feature, f, come in two broad forms: quantified and unquantified. Quantified generalisations speak of ‘all’ or ‘most’ or ‘few’ or ‘none’ or of some numerical fraction or particular percentage of Ks being or having f. Unquantified generalisations do not, though they may be expressed in singular or plural forms, in the first case with the definite article ‘the K is f’ in the second with the indefinite article ‘a K is f’, or simply ‘Ks are f’. The singular forms do not refer to individuals per se but to the kind, as when it is said that ‘the snake is a vertebrate’ or ‘a snake has scales’ and this sort of generic generalization is more familiarly expressed in the plural form as in ‘snakes are vertebrates’ or ‘snakes have scales’.
Being unquantified, these kinds of generalisations are not hostage to exceptions. So what is it that they capture? One idea would be observed prevalence among instances of the kind, but this is implausible in part for want of exposure to sufficient examples to conjecture prevalence. More plausibly they are expressions of identified characteristics attributed to the natures of the things in question. If these were essences they would not admit exceptions, but on the other hand if they are characteristics they are not entirely contingent either. What kind of thing then can generics be representing? The answer, I think, is propria: non-accidental, but nonessential characteristic of a kind of thing. Aquinas writes: ‘Substantial differences being unknown to us, or at least unnamed by us, it is sometimes necessary to use accidental differences in the place of substantial; as, for example, we may say that fire is a simple, hot, and dry body: for proper accidents are the effects of substantial forms, and make them known’.Footnote 41 Aquinian concepts, save in the case where what is up for conceptual characterisation is something whose real and nominal essence coincides, as in the example of geometrical figure or other broadly a priori knowable natures, are proprium-identifying generics that tolerate exceptions because propria unlike essences hold only for the most part ‘ut in pluribus’. We saw earlier, however, that natures as they exist in rebus are singular so if we are now to speak of general propria as expressed in unquantified generalisations we also need to recognize that these do not exist save as mind-dependent.
Lastly, as well those unquantified generalisations that attribute intrinsic characteristics, there are others that identify what look to be interest-reflecting features, such as ‘flames are hot’ or the comparative ‘tigers are more vicious than lions’ or, said of colours, ‘greens are more restful than reds’. Here, the point is that not only are general essences and propria not mind-independent in the metaphysical sense but the contents of many human concepts stand apart to some degree from the individual natures on the basis of which human concepts are formed. By ‘stand apart’, I mean that some at least of the concepts we form reflect our interests, and since the conceptual is always mind-dependent we can say that not only the fact but the structure and content of the general nature of things is a product of the human mind. This sounds a like the kind of thing Gilson was reacting against, but, if I am right, it is a position that a Thomist, to the extent that they are following the lead taken by Thomas himself, should accept. If by ‘the world’ one means the totality of particular things and particular characteristics, then the world may not be mind-dependent; but if one means by it the totality of existing things both particular and general, then in that sense both the generality and the inventory are mind-dependent. So, if mind-dependence is the mark of one kind of anti-realism, then Aquinas is a kind of anti-realist. If that seems unsatisfactory or misleading, then rather than challenge the substantive view so as to save him from that classification one might do better to recall the advice to avoid using the terminology of realism and anti-realism and attend instead to the particularities of what Aquinas has to say.