The Problem for Empiricists
Scientific knowledge of physical reality, at least in the strict, traditional sense, has to be based on necessary, universal generalizations. Such generalizations, in turn, are impossible without universal, substantial concepts of extramental objects. For if none of our concepts represents substantial characteristics of extramental objects, then our concepts can only represent contingent features of their objects, and so they cannot provide us with universal, necessary knowledge of these objects. It is therefore crucial to any epistemology upholding the possibility of scientific knowledge in this sense to account for the human mind’s ability to acquire substantial concepts of things in extramental reality.[1]
Such an account, however, is particularly problematic for empiricists, who, for the purposes of this paper, will broadly be characterized as philosophers holding that the human mind begins its existence in this life without any definite mental contents about extramental reality, in short, without any categorematic concepts,[2] but has to acquire its concepts in a natural process from experience.
To be sure, this is a somewhat broad characterization, which will turn out to comprise philosophers whom we usually would not subsume under the label ‘empiricist’ without reservation.[3] Nevertheless, this characterization certainly distinguishes a number of medieval Aristotelians from Platonists, Augustinians, and Cartesians, who would hold that the human mind begins its existence in this life in possession of at least some categorematic concepts, which therefore it does not acquire in this life from experience. Consequently, for these philosophers the acquisition of these concepts in this life is not a problem at all. They rather have trouble with accounting for the apparent lack of these concepts in children and mentally impaired adults, as well as the apparently mysterious match between these prenatal or innate concepts and the objects of empirical reality, and, in general, the supernatural dependency of what appears to be a natural operation of the human mind, namely, understanding.[4]
By contrast, the acquisition of substantial concepts in this life is a problem for empiricists, for they have to be able to show that these concepts can somehow be derived from the natural input the mind receives in this life, namely, sensory experience. However, sensory experience apparently can only provide the mind with information about sensible qualities of objects of experience, which are all accidental, non-substantial features of these objects. To be sure, if substantial concepts can be derived as some sorts of combinations of the concepts of these sensible qualities, then the problem may seem to be solved, in the way proposed by the British empiricists, Locke, Berkeley, and Hume. However, as John Buridan’s argumentation in q. 4 of bk. 1 of his Questions on Aristotle’s Physics had showed centuries before the British empiricist approach emerged, such a derivation is impossible.
In this paper, I will first argue, rather anachronistically, that Buridan’s discussion in this question amounts to a principled refutation of the British empiricists’ conception of our substantial concepts as “collections of simple ideas of sensible qualities”.[5]
After identifying the principles that allow this refutation, I will show that on the basis of two further Aristotelian principles Buridan can successfully defend the possibility of scientific knowledge of physical reality, while staying within the bounds of the broadly interpreted empiricism characterized above.
In conclusion, I will argue that it was precisely Buridan’s insistence on these Aristotelian principles that allowed him to be a thoroughgoing empiricist without slipping into the sort of “medieval Humeanism” from which he took great pains to distinguish his own nominalism, namely, the skepticism of Nicholas of Autrecourt and his ilk, that is, whomever else Buridan is opposing in this question.[6]
Buridan’s “Refutation of British Empiricism”
In his question-commentary on Aristotle’s Physics, Buridan raises the question “whether in every science the knowledge and understanding of things arises from the preexisting cognition of their causes, principles, and elements”.[7]
After advancing a number of arguments supporting the negative reply, Buridan begins his discussion by expounding an opinion on the issue, which he is about to refute:
“This question and the arguments brought up in connection with it raise several difficulties. One such difficulty is whether from the cognition (noticia) of one thing one can obtain the cognition of another. For there are two sorts of cognition, namely, complex and incomplex. About the incomplex sort some people say that no incomplex cognition can be obtained from another, since no cognition can be obtained from another, except by means of a consequence, but a consequence can only lead from a complex [cognition] to a complex one; therefore, etc. Second, they infer as a corollary that we have no cognition of any substance in terms of incomplex cognition, for we can arrive at the cognition of substances only by means of the cognition of accidents; and so by means of some consequence, which can only obtain between complex [cognitions]. But I do not agree with this opinion, and I posit two conclusions against it.”[8]
Buridan’s first conclusion directly attacks the first claim of this opinion, namely, that no simple cognition can be obtained from a simple cognition. He points out that the claim is self-defeating, insofar as the simple intellectual cognitions it involves had to come from some simple sensory cognitions, in line with the common assumption of the broadly understood empiricism described above; and so, some simple intellectual cognition had to come from some simple sensory cognition, whence some simple cognition had to be obtained from some simple cognition, contrary to the original claim. As he writes:
“The first [conclusion] is that some incomplex cognition can be obtained by means of another. For there are incomplex intellectual cognitions, and all intellectual cognitions are obtained by means of another act of cognition; therefore, some incomplex cognition is obtained by means of another [act of cognition]. The major premise has to be accepted, for if a caviler were to deny it, then [by virtue of this denial] he would have to concede at least the existence of some complex intellectual cognition; but the complex [cognition] would have to be composed of simple ones, for it is not divided to infinity as the continuum, and an intellectual cognition is not composed of sensory cognitions; therefore, it is composed of incomplex intellectual ones. But the minor of the principal argument is also clear, for at least the first intellectual cognition has to be obtained from a sensory one, and, in general, every intellectual cognition must be obtained from sensory cognition either directly or indirectly, since one who understands has to attend to [speculari] the phantasms, as is stated in book 3 of On the Soul; and for this reason it is also claimed in book 1 of the Posterior Analytics that if we lose one of our senses, we also lose the knowledge of the proper object of that sense.”[9]
So, simple intellectual cognition must somehow come from simple sensory cognition. But how is this possible? And even if we can provide an explanation of the derivation of simple intellectual cognition from simple sensory cognition in general, how do we know that we have such a simple cognition of substance obtainable from sense experience? Indeed, why would the intellectual cognition of substance have to be simple? After all, if the British empiricists are right, then the only way we can make sense of our substantial terms is to conceive of them as being associated with relatively stable collections of sensory ideas. The reason for this is that these terms certainly cannot be associated with anything over and above the sensory ideas we can gain from experience, whence they have to be associated with those relatively stable bundles of these ideas that the mind usually perceives together, and so associates them with substantive names, for practical reference.[10] Buridan’s second conclusion addresses this issue as follows:
“The second conclusion is that we have simple concepts of substances, for the concept of man from which we take the substantial term ‘man’ is a concept of substance, if man is a substance. And that concept supposits only for a substance, for if it supposited for an accident or for something composed from substance and accident, then it would not be true that man is a substance, for neither an accident nor something composed from substance and accident is a substance; but precisely a substance is a substance, and that concept, while it supposits for a substance, does not even connote an accident other than that substance, for then it would not belong to the category of substance, but to that of an accident, as do the terms ‘white’ or ‘big’ or ‘small’, etc. For these terms supposit for substance and not for anything else, just as the term ‘man’ does, but they leave the category of substance because of their connotation; therefore, a concept from which a term in the category of substance is taken is not a concept of any accident, or of something composed from substance and accident, but only of a substance or substances.
And if anyone were to say that they are complex, then the complex ones are combined from simple ones, for in the analysis of concepts one cannot go to infinity; and then those simple ones and the ones composed from them are only of substances; therefore, there are simple concepts of substances.”[11]
The first important thing to note about Buridan’s argumentation here is his insistence on the Aristotelian distinction between substance and accident, and his combination of this Aristotelian doctrine with his own semantic analysis of the terms and the corresponding concepts belonging to the Aristotelian categories. The point of the argument is that even if substances had complex concepts, those complex concepts would have to be made of simple concepts. But those simple concepts cannot be concepts of accidents, so those simple concepts would have to be simple substantial concepts, so we would still have to have some simple substantial concepts, which was the point to be proved. On the other hand, the claim that complex substantial concepts cannot be made up from accidental concepts (contrary to the British empiricists’ conception of collections of sensory ideas) is proved here with reference to Buridan’s doctrine of the semantics of substantial vs. accidental terms and concepts, as being absolute vs. connotative terms and concepts.
For Buridan, concrete substantial are distinguished from concrete accidental terms by their different modes of signification due to the different sorts of concepts to which they are subordinated, yielding their different modes of predication. Concrete substantial terms are subordinated to absolute concepts, whence they signify their significata absolutely, without relating them to anything else. Concrete accidental terms, on the other hand, are subordinated to connotative concepts,[12] whence they signify their significata in relation to their connotata, which are also called their appellata when they obliquely refer to these connotata in the context of a proposition. It is a consequence of this difference that substantial terms are predicated of their significata essentially or quidditatively, whereas the accidental terms are predicated of their significata non-essentially, or denominatively.[13] Accordingly, absolute terms, in particular substantial terms function in Buridan’s semantics as what we nowadays would call “rigid designators”. For these terms are true of their significata in a proposition as long as they supposit for them. But since the supposita of an absolute term are nothing but its significata that exist at the time connoted by the copula of the proposition in which the term is predicated, absolute terms always and necessarily supposit for their significata as long as these significata exist at the time connoted by the copula of the proposition in which they are predicated of these significata. So, these terms may never become false of these significata as long as these significata exist. Therefore, absolute terms are always predicated of their significata essentially, or quidditatively, and thus designate them “rigidly”. On the other hand, concrete accidental terms supposit for their significata only when their appellata belong to their significata in the way they are signified to belong to their significata. So, if the appellata cease to exist or cease to belong to the significata in the way demanded by the signification of the term, then these terms cease to supposit for their significata, whence they become false of their significata, even though these significata continue to exist. Therefore, the essential vs. non-essential predication of concrete substantial vs. accidental terms is a direct consequence of their mode of signification, which in turn, is determined by the sorts of concepts to which they are subordinated.
What Buridan’s argument shows is that the assumption that substantial concepts are collections of connotative concepts, which is precisely the implication of the British empiricist conception, would lead to the absurd conclusion that a substantial term would not be a substantial term, for then it would be subordinated to a non-substantial concept. As he writes further on:
“Again, if the substantial concept of man were complex, then let us posit that it consists of three simple ones, namely, a, b, and c. Then, if no concept of substance is simple, a can only be a concept of accident, and the same goes for b and c; therefore, the whole combined from them would also be only a concept of accident, and not one of substance, for a whole is nothing over and above its parts. But this is absurd, namely, that the substantial concept of man should be nothing but a concept of accidents; therefore, etc.”[14]
To be sure, the British empiricists, who provided precisely this sort of analysis for substantial terms, happily embraced this conclusion, and did not regard it as absurd at all. But Buridan’s previous argument, combined with his semantic considerations, also shows that this conclusion directly entails the impossibility of the essential predication of these “phony” substantial terms. This, however, entails further that they cannot serve as the basis for valid scientific generalizations: an implication that was to be worked out in the fullest detail by David Hume. But then, unless Humean skepticism is the inevitable consequence of empiricism in general, an empiricist who wants to save the possibility of scientific knowledge in the traditional sense has to be able to find an alternative way to account for the derivation of our substantial concepts from experience, without turning the terms associated with these concepts into non-essential predicates of their significata.
This is precisely what Buridan offers in his subsequent considerations, moderating his “empiricist nominalism” with “Aristotelian naturalism”, abandoned by his contemporary opponents, especially, Nicholas of Autrecourt.
Buridan’s Balancing Act: Empiricist Nominalism Combined with Aristotelian Naturalism
In response to the arguments supporting the opinion he rejects, Buridan offers four different ways in which one may account for obtaining some simple cognition from another without any inference. As he writes:
“Then, [I respond] to the arguments supporting this opinion. To the first, we have to reply that some cognition is obtained from another without inferring one proposition from another or others in four ways. First, objectively. For if there is some cognition in an external sense, then it is related to the cognition of the common sense as its object, and also any sensory cognition is related to intellective cognition as its object.”[15]
In this way, the higher cognitive faculty forms some act of cognition distinct from the act of cognition of a lower cognitive faculty, simply because it takes the act of the lower faculty as its object. To be sure, one has to make here the common distinction between an immediate and ultimate object: in the cognition of external objects (as opposed to the soul’s reflecting on its own acts) the act of the lower faculty is only the immediate object of the act of the higher faculty, insofar as the higher faculty cognizes the object of the lower faculty by means of cognizing the act of the lower faculty, in the same way as when I see my face in the mirror by means of its reflection. In any case, this certainly is the most general way in which one simple act of cognition can give rise to another, or indeed, in general, one stage of information processing can give rise to another, as when a picture taken by a digital camera is electromagnetically stored on a computer’s hard drive possibly for further processing. The important point here is that information received by one sort of encoder of that information can be actively used and further processed by another encoder, by reason of its own receptive and processing ability. In fact, in this way, the second encoder may even add information not contained in the first, as when a computer tags the picture files on its hard drive with time and date stamps. This is precisely the point Buridan makes concerning the second way in which a simple act of cognition may give rise to another:
“Second, [a simple act of cognition may give rise to another] elicitively, as Avicenna says that the estimative power from a sensed intention, namely, of color, or shape, or motion, elicits an intention not sensed, namely, that of attraction or repulsion [amicitie vel inimicitie]. This is why sheep fear and flee the wolf, and follow the shepherd. And this is not a miracle. Since the soul is much nobler than fire, yet fire in generating heat is able by that heat also to generate lightness and rarity, so it is reasonable that the soul, by means of one act of cognition is able to generate another one, naturally following upon the former.”[16]
This is indeed plausible; however, when he specifically addresses the issue of how simple substantial concepts may be derived from sensory cognition, Buridan warns us that this way of accounting for this specific process of concept acquisition may contain a false assumption. In his questions on Aristotle’s On the Soul, he analyzes the issue in the following way:
“… there is one way, in the first place, in which the cognition of accidents leads us to the cognition of substance. And this assumes first that the intellect is moved by phantasms, the imagination by the senses, and the senses by external objects. It assumes in the second place that the senses and the imagination are only of accidents. It assumes in the third place that the estimative power is superior to and more excellent than the external sensitive power; and so it is able to elicit from the sensed intentions some intentions not sensed. Thus also the intellect is superior to any sensitive power, whether external or internal; therefore, it is able from the intentions of accidents, which fell into the imagination elicit intentions of substances, which did not fall into the imagination. And so, by means of the cognition of accidents, we can arrive at the cognition of substances.
Briefly, this way [of addressing the issue] is defective in its second assumption, which was that the senses are only of accidents. For this goes against Aristotle, who in bk. 2 of this work [namely, On the Soul] asserts that the son of Diarus is sensed; although it is true that this is not per se, but per accidens. Indeed, we do not perceive substances under substantial concepts, but we do perceive them under accidental and connotative ones, and not under purely absolute ones.”[17]
So, even though the intellect may have the power to elicit intentions not contained in the senses, in the formation of substantial concepts it is simply not true that these would have to be “cooked up” by the intellect alone, for the sensory data provided by the senses about accidents does carry information about the substances to which these accidents belong. This is the idea that Buridan elaborates in the continuation of this passage, listing three further ways in which one can account for the intellect’s ability to form substantial concepts from sensory data, by extracting the information this sensory data carries about substances:
“The second way is that the senses first perceive both substance and accident in a confused manner, and afterwards the intellect, which is a superior power, differentiates between substance and accident. Therefore, if I see someone now to be white and later I see him to be black, and at the same time I perceive that he remains the same, I arrive at the cognition by which I notice that this is other than whiteness and likewise other than blackness. And thus, although at first substance and accident are apprehended by means of the senses in a confused manner, the intellect, which is a superior power, can arrive at the cognition of substance itself.
The third way is possible because things are cognized by means of their similitudes. For it is stated in bk. 3 of this work that “a stone is not in the soul, but the species of the stone is”. Since, therefore, it is the case that any effect bears the similitude of its cause, and an accident is an effect of a substance, it follows that an accident also bears a similitude of a substance, and consequently the intellect is able to arrive at the cognition of substance by means of the accident.
The fourth way can be this: prime matter, before a substantial form is educed from its potentiality, needs accidental dispositions preparing it for receiving such a form; the same can be imagined of the potential intellect, namely, that before there would be the similitude of substance in it, there have to be in it the species and similitudes of accidents. Once these are in the potential intellect, the agent intellect is able to extract from them the natural similitude of that substance to which those accidents belonged whose similitudes and intentions were in the potential intellect.”[18]
Basically the same point is made in the continuation of the previously discussed passage from the Physics-commentary.[19]
In view of these passages, we can summarize the Aristotelian principles allowing Buridan to maintain his empiricist nominalism without slipping into skepticism in the following way.
Once these two principles, which may be dubbed the principle of the activity of the intellect, and the principle of the substantial content of sensory information, respectively, are acknowledged, any empiricist should be able to provide a plausible account of our ability to acquire genuine substantial concepts from sensory information.[20] For in view of the first principle, the intellect is obviously able to extract content from sensory information which the senses could not so extract even though they may carry it, in the way, for instance, light received by a telescope carries not only visible information about the stars, but also information about their material constitution, which, however, is extractible only by means of spectral analysis. But in view of the second principle, the information about sensible accidents also carries such extractible information about the substances to which these accidents belong. Therefore, the intellect should be able to form genuine substantial concepts from this sensory information. But then, these genuine substantial concepts will be denoted by essential predicates of the things conceived by means of these concepts, which will always necessarily apply to these things as long as these things exist. And so, these predicates will be scientifically knowable characteristics of these things.
All in all, even if, perhaps, Nicholas of Autrecourt was “the medieval Hume”, it did not take a “medieval Kant”[21] to refute his skepticism. For Buridan’s version of an essentialist nominalism was sufficient to show that one can be a nominalist and a thoroughgoing empiricist without having to fall prey to any serious form of skepticism. In this way, Buridan’s essentialist nominalism could, in principle, have shown a way out of the dilemma of empiricism vs. rationalism of early modern philosophy. Indeed, the dilemma may not even have emerged in its original form, if the Aristotelian empiricism of the scholastics, including Buridan’s, had not been abandoned earlier, partly for extrinsic reasons, by the new intelligentsia of a new era.
Gyula Klima
Fordham University
[1] To be sure, our ability to acquire such concepts will still not guarantee that we know which of our concepts are the essential ones. That is the task of empirical research to find out. But we can know a priori that if we cannot have such concepts, then we cannot have scientific knowledge in the specified sense. For more on this issue see Klima, G.: “Contemporary ‘Essentialism’ vs. Aristotelian Essentialism”, forthcoming in: Haldane, J. (ed.): Mind, Metaphysics, and Value in the Thomistic and Analytic Traditions, University of Notre Dame Press: Notre Dame, IN, 2002.
[2] Obviously, syncategorematic concepts, such as the concepts of the Boolean operations of negation, conjunction, etc., may consistently be treated even by empiricists as innate operations of the mind, not carrying any information about extramental reality, but simply operating on categorematic concepts which do carry such information.
[3] For a discussion of the issue of ‘empiricism’ in late medieval philosophy, see Zupko, J.: “What Is the Science of the Soul? A Case Study in the Evolution of Late Medieval Natural Philosophy”, Synthese, 110(1997), pp. 297-334.
[4]
As Matthew of Aquasparta remarks in connection with the doctrine of divine
illumination: “…if that light were the entire and sole reason for
cognition, then the cognition of things in the Word would not differ from their
cognition in their proper kind, neither would the cognition of reason differ
from the cognition of revelation, nor philosophical cognition from prophetic
cognition, nor cognition by nature from cognition by grace.” Bonaventure, et al., De Humanae Cognitionis Ratione:
anecdota quaedam Seraphici Doctoris Sancti Bonaventurae et nonnulorum eius
discipulorum, Ad Claras Aquas (Quaracchi): St. Bonaventure, 1883, pp.
94-96.
[5] “I say, our specific ideas of substances are nothing else but a collection of a certain number of simple ideas, considered as united in one thing. These ideas of substances, though they are commonly simple apprehensions, and the names of them simple terms, yet in effect are complex and compounded. Thus the idea which an Englishman signifies by the name swan, is white colour, long neck, red beak, black legs, and whole feet, and all these of a certain size, with a power of swimming in the water, and making a certain kind of noise, and perhaps, to a man who has long observed this kind of birds, some other properties: which all terminate in sensible simple ideas, all united in one common subject.” John Locke: An Essay Concerning Human Understanding (Essay), New York: Dover Publications, 1959, bk. II, c. 23, para. 14.
[6] See Hastings Rashdall, “Nicholas de Ultricuria, a Medieval Hume”, Proceedings of the Aristotelian Society 8(1907), pp. 1-27, and T.K. Scott, “Nicholas of Autrecourt, Buridan and Ockhamism”, Journal of the History of Philosophy 9(1971), pp.15-41. In his “John Buridan and Nicholas of Autrecourt on Causality and Induction”, in Traditio, 43(1987), pp. 237-255, Hans Thijssen has plausibly argued that since some of the theses and arguments Buridan opposes here do not reflect Nicholas’ doctrine as we know it, Buridan may well have had other opponents in mind. On the other hand, since the theses and arguments in question are at least not incompatible with Autrecourt’s known doctrines, it is still possible that Buridan had in mind some further works or even just oral presentations of Autrecourt’s that we simply do not know of from other sources. Indeed, this latter alternative has the advantage of explaining the phenomena per pauciora. In any case, my subsequent argument is not dependent on the identity of Buridan’ actual target of criticism in this question.
[7] Buridan: Quaestiones super Octo Libros Physicorum Aristotelis (QiP), Minerva: Frankfurt A.M., 1964, lb. 1, q. 4.
[8] Ibid.
[9] Ibid.
[10] See Locke, Essay, bk. II. cc. 22-23; bk. III. cc. 5-6; George Berkeley, A Treatise Concerning the Principles of Human Knowledge (Principles), Hackett: Indianapolis, 1982, nn. 1, 24, 54; David Hume, A Treatise of Human Nature (Treatise), Oxford: Clarendon Press, 1978, Bk. I, sect. VI.
[11] QiP, lb. 1, q. 4.
[12] John Buridan: Summulae de Dialectica (Summulae), New Haven: Yale University Press, 2001, pp.147, 173, 639, 642, 644-646, 729, 735.
[13] Summulae, pp.106, 123, 126-128, 131, 135, 138, 147-149, 147n9, 155, 155n20, 156-158, 163, 169, 169n38, 175, 183, 202, 629, 640, 653, 668, 732, 787, 885, 886.
[14] QiP, lb. 1, q. 4
[15] QiP, lb. 1, q. 4
[16] Ibid.
[17] J. Buridan: Le traité de l’âme de Jean Buridan (De Prima Lectura). Édition, Étude critique et doctrinale par B. Patar, Éditions de l'Institut Supérieur de Philosophie: Louvain-la-Neuve, 1991, Quaestiones in De Anima (QDA), lb. 1, q. 5. Cf. “Hence it is plain we do not see a man - if by man is meant that which lives, moves, perceives, and thinks as we do - but only such a certain collection of ideas as directs us to think there is a distinct principle of thought and motion, like to ourselves, accompanying and represented by it.” Berkeley, Treatise, Part I, n. 148, p. 88.
[18] QDA, lb. 1, q. 5. (prima lectura)
[19] “Third, abstractively, as when I first have a concept that represents substance and accident together in a confused manner, for example, when I perceive something white, for I see not only whiteness, but something that is white, and then if I perceive the same thing to move and change from white to black, then I judge that this is something distinct from whiteness, and then the intellect naturally has the power to analyze that confusion, and to understand substance abstractively from accident, and accident abstractively from substance, and it can form a simple concept of each, and it is in the same way, by abstraction, that a universal concept is formed from a singular one, as one should see in bk. 3 of On the Soul, and bk. 7 of the Metaphysics.” QiP lb. 1, q. 4.
[20] To be sure, one might still raise the question whether Buridan is “entitled” to these Aristotelian principles in his solution, given his semantic ideas concerning the natural signification of absolute concepts. But this question is beyond the scope of the present paper. Cf. P. King: “John Buridan’s Solution to the Problem of Universals”, in: J.M.M.H. Thijssen and J. Zupko (eds.): The Metaphysics and Natural Philosophy of John Buridan, Brill Publishers: Leiden, 2001, pp. 1-27.
[21] It may be interesting to note here that from a medieval perspective Kant’s solution to “Hume’s problem” may be characterized as simply shifting the categorical status of some fundamental metaphysical concepts, such as ‘substance’, ‘accident’, ‘cause’, ‘effect’, ‘existence’, etc. Instead of treating them as (whether innate, infused, or empirically acquired) categorematic concepts, he treats them as “logical functions”, i.e., syncategorematic concepts. But then it is no wonder that in the conceptual framework of post-Kantian positivist philosophy a number of traditional metaphysical problems will turn out to be not only radically undecidable, but even meaningless, containing “category mistakes”. At the meeting Stephen Read has also called my attention to Thomas Reid’s very different, “common sense” criticism of Hume’s philosophy, as bearing some remarkable resemblances to Buridan’s approach to the issue. In fact, there may even be some actual historical connection between their ideas, given the lasting influence of Buridan’s thought in Scotland through the circle of John Major. Cf. J. Haldane, “Reid, Scholasticism and Current Philosophy of Mind”, in: M. Dalgarno and E. Matthews (eds.): The Philosophy of Thomas Reid, Kluwer Academic Publishers: Dordrecht, 1989, pp. 285-304.