The Kantian thing-in-itself is nothing but a point of view : a critique of Searle’s external realism

The three last chapters of John R. Searle’s The Construction of Social Reality are devoted to the defence of the two presuppositions which support the rest of the book : (i) the idea that there is an independent reality (external realism, noted ER) and (ii) a defence of the correspondence theory of truth. Here, we will consider the first point.
The external realism defence in chapters VII and VIII is presented with pedagogy and clarity. The author begins by denouncing the usual confusions about realism in a “the realism is not…” chain which is so long that it brings forth an inevitably minimalistic conception of realism – so minimalistic one would wonder what it still can be. Here, we shall focus on the second of these confusions : “Another misconception is to suppose that there is something epistemic about realism” [1].
In this paragraph of approximately twenty lines, Searle envisages refuting this H. Putnam’s statement : “the whole content of Realism lies in the claim that it makes sense to think of a God’s Eye View (or better a view from nowhere)” [2]. Here is his answer :

On the contrary, the whole idea of a « view » is already epistemic and ER is not epistemic. (1) It would be consistent with realism to suppose that any kind of “view” of reality is quite impossible. (2) Indeed, on one interpretation, Kant’s doctrine of things in themselves is a conception of a reality that is inaccessible to any “view”. [3]

This answer is disappointing on two counts. In accordance with the enumeration above, I shall analyze it in two moments.

I- Searle vs. Putnam : the view from nowhere

First of all, let us examine the statement that “It would be consistent with realism to suppose that any kind of “view” of reality is quite impossible”. It is obvious, and even coarse, considering the given definition of ER. Few pages before, ER was defined as the doctrine assuming that “reality exists independently of our representations of it” [4]. The author would certainly agree to say “independently of any kind of representation”, as to recognize that the propositions “the representation of reality” and “the view of reality” are equivalent. So, the formal definition could be expressed as follows : ER is the doctrine assuming that reality exists independently of “any kind of ‘view’ of reality”.
Now, let us call X the proposition “any kind of ‘view’ of reality”. What Searle is saying in his answer to Putnam is nothing but assuming that reality exists independently of X would be perfectly compatible with supposing that X is impossible. Yet, by definition, saying that reality exists independently of X is saying that it is not dependent on its possibility, therefore it would not be altered by its impossibility. Thus, Searle’s answer is already contained in his definition of ER.
Next, is it really an answer ? Putnam accuses realism for being dependent on a view from nowhere [5]. Searle defends that realism is independent of any kind of view – implicitly “including the view from nowhere”. As we just showed, this statement adds nothing to realism ; to be independent of any kind of view is nothing but its definition. In other words, Putnam actually had this definition in mind when he wrote his accusation. Then, we arrive at the following conclusion : when Putnam accuses realism to be dependent of a view from nowhere, he precisely accuses it of pretending to be independent from any kind of view. So, Searle does not answer the accusation ; he confirms it – he drives the nail home. The problem comes from having not sufficiently thought about the notion of “view” or “point of view” from nowhere. A point of view from X is a view from the point X. A point of view from nowhere, therefore, is a view from the point “nowhere”, that is to say from no points, no points of view. Then, Putnam’s “view from nowhere” and Searle’s “no point of view” are only one and the same thing.
This discloses Searle’s real presupposition, which is that the realism’s point of view from nowhere is simply not a “point of view”, as it is the point of view of no subject. In short, it is the object point of view [6]. In a previous passage on the definition of knowledge, which the author accepts in the traditional and debatable sense of “Justified True Belief”, Searle gave an implicit definition of objectivity : “Knowledge is thus by definition objective in the epistemic sense, because the criteria for knowledge are not arbitrary, and they are impersonal” [7]. Objective is what is not arbitrary and is also impersonal. This second criterion draws my attention. Are the criteria for knowledge im-personal ? Are they the work of no one ? Have you ever encountered knowledge which is affected by no one ? Strictly speaking, it seems to me that knowledge criteria are not im-personal (no one’s point of view, objectivity) but inter-personal : convergent points of view of the majority, inter-subjectivity.
Let us recapitulate. When Searle claims that its ER is independent from any kind of point of view, we should read “point of view from somewhere”, that is to say “subjective point of view”. Now, by writing that, he implies immediately that his ER is dependent of the “point of view from nowhere”, that is to say “objective point of view”. First conclusion : Putnam is absolutely right. Realism is entirely based on the conviction that it makes sense to think of a view from nowhere – or, in Searle’s words, that it makes sense not to think of a view from somewhere, which amounts to exactly the same thing. Second conclusion : Searle’s ER is dependent of the object point of view, and as such he refers to the Kantian Ding an sich in his second part. Let us examine it now.

II- Searle vs. Kant : the real nature of the Kantian

The Kantian thing-in-itself doctrine has been debated since 1787 [8]. In more than two centuries, it has been understood only rarely. The vast majority of readers still think today that the thing-in-itself embodies objectivity, that is to say it is a distinct and above all real object – even more real than the rest. This is the way that Searle leans on it – like the pure object, independent from the subject, as “a conception of a reality that is inaccessible to any view”. But this prejudice is absolutely false. The Kantian thing-in-itself doctrine is particularly misunderstood and requires rediscovery.

1. The distinction between the thing-in-itself and the phenomena is not an objective distinction between two objects, but a subjective one between two points of view

First of all, one should be conscious that the thing-in-itself is only a correlative and negative concept ; that is to say, it cannot be apprehended independently of its correlatum : the phenomena. Kant writes : “The correlatum of the thing in the phenomena is the thing-in-itself (…). The concept of a thing-in-itself (ens per se) springs only from an earlier given concept, the concept of the object in the phenomena ; it springs from a relation where the object is considered relatively and to tell the truth in a negative relation” [9].
Therefore, a good “thing-in-itself” study should first be interested in what distinguishes it from the phenomena. What does Kant say about this distinction ? He writes it is not objective but only subjective : “the distinction of the conceptions of a thing-in-itself and of one in appearance is not objective but subjective only [nicht objectiv sondern blos subjectiv]” [10]. He writes it is not real but only ideal : “the objects are representations in appearance, and their difference from things-in-themselves is not a difference of the objects as things but is only a scientific (ideal) difference for the subject, not one for the object [ein scientisischer (ideal) für das Subject nicht das Object]” [11]. In other words, there are not two different objects but only one : it is the same “something” which is phenomenon and also thing-in-itself. The thing-in-itself and the phenomena are ontologically identical.
Consequently, what about the distinction, if it concerns the same object ? It can be nothing but subjective, between two points of view [12]. The thing-in-itself is not another object, but only another point of view of the same object : “The thing-in-itself = X does not mean another object, but only another point of view [Standpunct], negative, from where precisely the same object is considered” [13]. It is only another mode of representation : “The object in itself = x is the sense-object in itself, but as another mode of representation [Vorstellungsart], not as another object” [14]. In the Critique of Pure Reason, Kant is talking about this distinction as one between “two different standpoints”, “points of view” [15] or even two “significances” (Bedeutungen) : the Critique of Pure Reason “has not erred in teaching that the object should be taken in a twofold meaning, namely as appearance or as thing in itself” [16].
A point of view is a view from a point. Consequently, if the thing-in-itself and the phenomena are two different views of the same object, that means they originate from two different points, and here lies the distinction. Kant is very explicit : the phenomena point of view is the one of intuition, while the thing-in-itself point of view is the one of reason. The distinction is nothing but the two sides of appearance : “appearance, which always has two sides, one where the object is considered in itself (without regard to the way in which it is to be intuited, the constitution of which however must for that very reason always remain problematic), the other where the form of the intuition of this object is considered” [17]. Intuition considers things as phenomena and reason considers them in themselves, “i.e., without taking account of the constitution of our sensibility” [18].
Let us sum up. The distinction between the thing-in-itself and the phenomena is not an objective distinction between two different objects, but a subjective one between two points of view or two modes of representation : the one of reason and the one of intuition.

2. The thing-in-itself is therefore a point of view, a thought-object (ens rationis), which does not exist outside of the understanding

Saying that the thing-in-itself is nothing but a point of view, a mode of representation, is to say it is an object only as a thought-object. On many occasions, Kant characterizes it as an ens rationis : “The thing in itself is a thought-object [Gedankending], (ens rationis) of the connection of this manifold whole into the unity to which the subject constitutes itself” [19] ; “the object as thing in itself, which is only an ens rationis (that is, only thought-object) [welches nur ens rationis d. i. nur Gedankending] and, determining, not objectively but only subjectively, is a conceptus infinitus (indefinitus)” [20]. He also writes that the thing-it-itself is nothing but the simple “position of a thing of thought [die Position eines Gedankendinges]” [21].
Elsewhere, Kant refines the thing-in-itself as nothing but a principle : “the thing in itself is not an existing being but = x, merely a principle [bloss ein Princip ist]” [22] ; “The thing in itself = x is not an object given to the senses, but only the principle of synthetic a priori knowledge of the manifold of sensible intuition in general, and of the law of its coordination” [23] ; “The distinction of the so-called thing in itself [des sogenannten Gegenstandes an sich], in contrast with the thing in the phenomena (phenomenon adversus noumenon) does not mean that there is a real thing which contrasts with the physical thing, but it points out that x is only the principle that there is nothing empirical in that which holds the determining foundation of the possibility of experience” [24].
Then, when we are talking about the Kantian thing-in-itself, we are talking about nothing but a certain manner of considering things. Too often, one forgets that in the texts, things-in-themselves are nothing but things considered or regarded as such : Kant does not talk about “2” but only about “Ding an sich selbst” – the correct and complete expression being even “Ding an sich selbst betrachtet” [25]. This confusion in terms, this intellectual laziness which persists in keeping only the “Ding an sich” while forgetting the “selbst betrachtet”, is probably the major cause of the misunderstanding of the Kantian thing-in-itself doctrine. By using “2” out of context, the commentators and, consequently, the readers end up thinking that the Kantian thing-in-itself, as the name indicates, is well and truly a “thing”. One could justifiably object that the Kantian text is more ambiguous than it appears and that the author let himself go often to this deceptive short cut of writing “Ding an sich” rather than the more precise “Ding an sich selbst betrachtet”. This is true, but Kant knew what he was talking about ; he was conscious that his thing-in-itself, shortened for the use, was only an as if (als ob), an as if the thing were in itself – just a figure of speech [26].
Point of view, mode of representation, thought-object, principle… The thing-in-itself is finally nothing but an act of the understanding : “the intelligible object is not an objectum noumenon, but the act of the understanding [Nicht objectum noumenon, sondern der Akt des Verstandes] which constitutes the object of the sensory intuition as simple phenomena” [27]. In other words, and this conclusion should be underlined by the realistic readers, the thing-in-itself has no external existential value, it does not exist outside the understanding. To the phenomenon “corresponds its opposite, the noumena, not as a particular thing but as an act of the understanding = x [als Akt [= Gedanke] des Verstandes = x], act which is absolutely nothing outside of the understanding other than an object in general and which is only in the subject himself” [28].
More generally, we can conclude that if the thing-in-itself does not exist outside of the understanding, that is to say outside of the subject, the subject is the thing-in-itself – in a way – and this way is the reason. “The subject is the thing-in-itself because it holds a spontaneity : the phenomena is receptivity” [29]. Let us finally remember this last definition which summarizes all the rest : “the thing-in-itself ; it is the subject that I make object [das Ding an sich, ist das Subjekt, welches ich zum Objekt mache]” [30]. The thing-in-itself is nothing but the subject itself, a constructive subject which turns itself into an object [31].
Let us review. The thing-in-itself, or more exactly the thing considered in itself (Ding an sich selbst betrachtet), is nothing but a point of view, a mode of representation, a thought-object (ens rationis), a principle, a certain way to consider things, an act of the understanding, which does not exist outside of the subject, and which is nothing other than the subject making itself an object.

3. Answer to an objection

Before concluding, we have to anticipate an obvious objection that the reader who is familiar with the vast secondary literature about the Kantian thing-in-itself doctrine would be tempted to raise. Here is the objection :

Your short presentation of the Ding an sich is partial, therefore biased. In fact, there are two possible interpretations of the Kantian doctrine, and Kant himself oscillates between both positions : the idealist version that you are defending, which is found nowhere else but in the Opus Postumum (a jumble of disjointed reflections written by the later Kant and published without the author’s permission), and the realistic version, supported by some more serious and reliable texts which assume the existence of the thing-in-itself.

It seems to me that when one talks about two possible interpretations – realistic and idealistic – of the Kantian thing-in-itself, we are misunderstanding Kant, who actually states that the thing-in-itself exists. However, it is not a realistic interpretation to say that the thing-in-itself exists. It would be one to say that this thing-in-itself is an object independent of the subject – but Kant does not say that to my knowledge. There are two questions that it would be proper not to confuse : the one, ontologically first, of the existence of the thing-in-itself (Is there any thing-in-itself ?) and the one, ontologically second, of its nature (what is this thing-in-itself that exists ?). If, in this paper, I only claim to answer the second question, it is because I think the first raises no difficulty : yes, the thing-in-itself exists in Kant, it would be absurd to deny it [32]. The only true challenge does not concern its existence, but its nature.
Here, an intermediate objection would be : “if the thing-in-itself is unknowable, what can we know about its nature ?” Sophism. Strictly speaking, it is not the thing-in-itself, as a genus, which is unknowable, but the particular things-in-themselves : I can’t know this desk in itself, this pen in itself, air in itself, ice in itself. But I can assuredly know what it is to be in itself. The question of the nature of the thing-in-itself, therefore, is not the one of its content (which would be asked only about particular things) but the one of its status, more precisely of its status relative to the subject.
Let us conclude. I absolutely do not dispute the existence of the thing-in-itself in the Kantian doctrine. But to affirm the existence of the thing-in-itself is not sufficient to constitute a realistic interpretation : even then the thing-in-itself should be independent of the subject, which is absolutely not the case. Therefore, I question the thesis that there are two possible interpretations, one realistic and the other idealist, of the Kantian doctrine of the thing-in-itself. The alleged realistic interpretation is either not realistic, if it consists in nothing but asserting the existence of the thing-in-itself, or it is false, if it asserts its independence from the subject. Consequently, the so-called “idealist interpretation” is not an interpretation : it is the Kantian doctrine itself. And instead of idealism, an ambiguous and historically loaded term, I prefer talking about antirealism.

Conclusion : the elimination of a realistic illusion

Searle wrote : “Kant’s doctrine of things in themselves is a conception of a reality that is inaccessible to any ‘view’”. [33] What should we answer ? First of all, the Kantian doctrine of things in themselves is certainly not a conception of a “reality”. Kant himself reminds us that the concept of “reality” can on no account be applied to the thing-in-itself because it applies only to the sensible domain (space and time), that is to say to the phenomena : “the concepts of reality, substance, causality, indeed even necessity in existence, lose all meaning and are empty titles for concepts without any content when with them I venture outside the field of sense.” [34] From a Kantian point of view, the thing-in-itself reality is nothing other than a contradiction in terms, an absolute non-sense.
Then, the Kantian thing-in-itself is not “inaccessible to any view” for two reasons : on one hand, as thought-object (ens rationis), the thing-in-itself not only can but even should be thought. The thing-in-itself is only inaccessible to knowledge, but it is by definition accessible to thought. On the other hand, as we saw in the first part, to say it is “inaccessible to any view” comes down to say it is the object point of view, inaccessible to the subject. But as we saw in the second part, the thing-in-itself is itself a point of view of the subject ; it is the subject making of itself an object.
The realistic reading that leans on the Kantian doctrine of the thing-in-itself as a paradigm of objectivity is therefore a complete misinterpretation. This reading, sadly very frequent, testifies of a serious misunderstanding of Kant [35]. If we examine the texts carefully, the Kantian doctrine of the thing-in-itself even appears as one of the strongest manifestation of Kantian antirealism and constructivism.
To realistic readers, therefore, we suggest two things. Firstly, that they accept their necessary dependence on a mysterious “point of view from nowhere”, which can be expressed, as in the case with Searle, like a negation of the points of view from somewhere, which comes strictly to the same. Secondly, that they abandon the habit to use the Kantian doctrine of the thing-in-itself to support their thesis, because in so doing, they commit a grotesque misinterpretation.

[1] J. R. Searle, The Constrution of Social Reality (New York : Free Press, 1995), 154.

[2] H. Putnam, Realism with a Human Face (Cambridge, MA : Harvard University Press, 1990), 23.

[3] J. R. Searle, ibid. I add (1) and (2).

[4] J. R. Searle, op. cit., 150.

[5] According to Putnam’s intention, I favour “from nowhere” rather than “divine”.

[6] A few pages after, Searle writes : “In short, it is only from a point of view that we represent reality, but ontologically objective reality does not have a point of view” (op. cit., 176).

[7] Searle, op. cit., 151.

[8] It seems that Jacobi was the first to launch the offensive, in the appendice “Beilage über den transscendentalen Idealismus” of his book David Hume über den Glauben oder Idealismus und Realismus (1786).

[9] Immanuel Kant, Kants Opus Postumum dargestellt und beurteilt, ed. E. Adickes, Kant-Studien, Ergänzungshefte im Auftrag der Kant-Gesellschaft, n. 50 (Berlin : Verlag von Reuther & Reichard, 1920), A 571, 652-653, my translation.

[10] Immanuel Kant, Opus Postumum, AK XXII 26, my translation.

[11] Immanuel Kant, Opus Postumum, AK XXII 74, my translation.

[12] See W. H. Werkmeister, “The Complementarity of Phenomena and Things-in-Themselves”, Synthese 47 : 301-312. Reprinted in Ruth F. Chadwick and Clive Cazeaux, eds., Immanuel Kant. Critical Assessments, volume II (London and New York : Routledge, 1992), 276-285. He is one of these rare commentators who saw the subtlety. He writes : “the distinction between objects of experience as phenomena and as things-in-themselves is not an ontological distinction but one of the perspective for viewing the objects” (278).

[13] Immanuel Kant, Opus Postumum, AK XXII 42, my translation.

[14] Immanuel Kant, Opus Postumum, AK XXII 414, ed. Eckart Förster, trans. Eckart Förster and Michael Rosen (Cambridge : Cambridge University Press, 1993), 181. See also : « Dieser Unterschied liegt nicht in den Objecten sondern blos in der Verschiedenheit des Verhältnisses wie das den Sinengegenstand apprehendirende Subject zur Bewirtung der Vorstellung in ihm afficirt wird » (Opus postumum, AK XXII 43).

[15] “thus to experiment will be feasible only with concepts and principles that we assume a priori by arranging the latter so that the same objects can be considered from two different sides, on the one side as objects of the senses and the understanding for experience, and on the other side as objects that are merely thought at most for isolated reason striving beyond the bounds of experience. If we now find that there is agreement with the principle of pure reason when things are considered from this twofold standpoint, but that an unavoidable conflict of reason with itself arises with a single standpoint, then the experiment decides for the correctness of that distinction” (Immanuel Kant, Critique of Pure Reason, B XVIII, ed. Paul Guyer & Allen W. Wood (Cambridge : Cambridge University Press, 1998), 111, note).

[16] Immanuel Kant, Critique of Pure Reason, B XXVII, op. cit., 116.

[17] Immanuel Kant, Critique of Pure Reason, A 38 / B 55, AK III 63, op. cit., 183.

[18] Immanuel Kant, Critique of Pure Reason, A 28 / B 44, AK III 56, op. cit., 160.

[19] Immanuel Kant, Opus Postumum, AK XXII 414, ed. Eckart Förster, op. cit., 180-181.

[20] Immanuel Kant, Opus Postumum, AK XXII 420, ed. Eckart Förster, op. cit., 184.

[21] Immanuel Kant, Kants Opus postumum, ed. E. Adickes, op. cit., C555, 694, my translation. More generally, we refer to p. 669-689 for the thing-in-itself as “transzendentaler Gegenstand = x” (669).

[22] Immanuel Kant, Opus Postumum, AK XXII 34, ed. Eckart Förster, op. cit., 175.

[23] Immanuel Kant, Opus Postumum, AK XXII 33, ed. Eckart Förster, op. cit., 175.

[24] Immanuel Kant, Kants Opus postumum, ed. E. Adickes, op. cit., C549, 670, my translation.

[25] Here are some examples which could easily be multiplied : “if I consider all things not as phenomena but rather as things in themselves [sondern als Dinge an sich betrachte] and as objects of mere understanding” (Critique of Pure Reason, A 206 / B 251, AK III 177, op. cit., 314, underlined by me) ; “it [our understanding] is not limited by sensibility, but rather limits it by calling things in themselves (not considered as appearances) [(nicht als Erscheinungen betrachtet)] noumena” (Critique of Pure Reason, A 256 / B 312 AK III 212, op. cit., 351, underlined by me) ; “The word absolute is now more often used merely to indicate that something is valid of a thing considered in itself [an sich selbst] and thus internally” (Critique of Pure Reason, A 324 / B 381, AK III 252, op. cit., 401, underlined by me) ; “the laws of nature can never be cognized a priori in objects insofar as these objects are considered [sondern als Dinge an sich selbst betrachtet werden] not in relation to possible experience, but as things in themselves” (Prolegomena to Any Future Metaphysics, AK IV 296, ed. Gary Hatfield, Cambridge, MA : Cambridge University Press, 1997, 49, underlined by me).

[26] “Now this being of reason (ens rationis ratiocinatae) is, to be sure, a mere idea, and is therefore not assumed absolutely and in itself as something actual, but is rather taken as a ground only problematically (because we cannot reach it through any concepts of the understanding), so as to regard all the connection of things in the world of sense as if they had their ground in this being of reason, but solely with the intention of grounding on it the systematic unity that is indispensable to reason and conducive in every way to empirical cognition of the understanding but can never be obstructive to it” (Critique of Pure reason, A 681 / B 709, AK III 449, op. cit., 611).

[27] Immanuel Kant, Kants Opus postumum, ed. E. Adickes, op. cit., A573, 650, my translation.

[28] Immanuel Kant, Kants Opus postumum, ed. E. Adickes, op. cit., C599, 674, my translation.

[29] Immanuel Kant, Kants Opus postumum, ed. E. Adickes, op. cit., A573, 650, my translation. See Pierre Lachièze-Rey, L’idéalisme kantien (Paris : Vrin, 1972), 183 : “le sujet est immédiatement identifié avec la chose en soi (…) [cette identification est] destinée à idéaliser la chose, ou, tout au moins, à combattre toute théorie qui la traiterait comme une réalité indépendante de l’activité spirituelle”.

[30] Immanuel Kant, Kants Opus postumum, ed. E. Adickes, op. cit., A571, 652, my translation.

[31] See Roger Daval, La métaphysique de Kant. Perspectives sur la métaphysique de Kant d’après la théorie du schématisme (Paris : PUF, 1951), 304 : “l’objet en soi émane d’un sujet constructeur”.

[32] Kant writes that if “it would be absurd for us to hope that we can know more of any object than belongs to the possible experience of it (…), it would be on the other hand a still greater absurdity if we conceded no things in themselves” (Prolegomena, §57, AK IV 350-351, op. cit., 109).

[33] J. R. Searle, op.cit., 151.

[34] Immanuel Kant, Critique of Pure Reason, A 679 / B 707, AK III 447-448, op. cit., 610. See also : “The understanding (…) thinks of an object in itself, but only as a transcendental object (…) that cannot be thought of either as magnitude or as reality or as substance, etc. (since these concepts always require sensible forms in which they determine an object)” (A 288 / B 344, AK III 231, op. cit., p. 381).

[35] Searle commits another mistake about the Kantian Ding an sich when he writes it is unthinkable. He writes the Kantian Ding an sich is “beyond the grasp not only of our knowledge but of our language and thought” (op. cit., 174), which is inexact. The Kantian Ding an sich is unknowable, but it’s necessarily thinkable, as thought-object, ens rationis. Let us notice, by the way, that Searle’s distorting realistic reading is all the more inconsistent since he seems not to ignore himself to be dealing with a transcendental idealist : “Kant’s transcendental idealism is a more sophisticated variant of it than one finds in Berkeley” (168).