A transcendental argument, simply defined, is an argument purporting to demonstrate that some X (such as a particular concept, belief, or state of affairs) is a necessary precondition of some undeniable feature of human cognition (e.g., that we have orderly experiences or make judgments). At least, that is the conventional understanding of the term today. But who was the first to use the term in that sense?
Surprisingly, it might have been Cornelius Van Til.
I’m currently working on a monograph on Van Til’s transcendental argument for God (more precisely, for Christian theism). Preparatory research has required me to review everything Van Til says, explicitly or implicitly, about transcendental argumentation across his corpus, and to take a deep dive into the contemporary literature on transcendental arguments. In the process, I discovered something quite interesting.
Immanuel Kant is rightly credited with bringing to prominence the idea of a transcendental argument and deploying several instances of transcendental argument. However, while he uses the phrases “transcendental argument” and “transcendental proof” in his Critique of Pure Reason (more precisely, those phrases appear in the English translations of CPR), he isn’t referring in those places to what are now referred to by those terms (or to what are now considered to be Kant’s main transcendental arguments). Indeed, there’s only one instance of “transcendental argument” in CPR (at least in the Kemp Smith and Guyer-Wood translations: A627/B655) and in that context it refers to an argument with a conclusion that transcends the limits of human reason (specifically, a classical theistic argument).
So who was the first to use “transcendental argument” in the modern sense?
Barry Stroud, a leading authority on TAs, wrote the following in a 1991 essay:
The earliest use of the phrase “transcendental argument” that I am familiar with is by J. L. Austin in a 1939 Joint Session symposium on ‘Are There A Priori Concepts?’ As far as I know, Kant never used the corresponding German expression, although he spoke of many different things as ‘transcendental’, including ‘transcendental philosophy’. I have not seen the expression “transcendental argument” used anywhere in the nineteenth century, or in the twentieth before Austin. I cannot say I have searched.
Paul Franks, writing in the same volume, concurs, albeit with some hedging:
The earliest instance of current usage I have found is in Austin . . . in a paper given in 1939. Since Austin does not give the impression of introducing a usage, it is likely that earlier instances are to be found.
Adrian Bardon, another expert on TAs, writes:
“Transcendental” reasoning, for Kant, is reasoning pertaining to the necessary conditions of experience. Though he did coin the term “transcendental argument” in a different context, Kant actually did not use it to refer to transcendental arguments as they are understood today. He did sometimes use the term “transcendental deduction” for a range of arguments concerning the necessary conditions of coherent experience. Early uses of the term “transcendental argument” for arguments of this type have been noted in Charles Peirce and J. L. Austin.
The sources Bardon cites are dated 1931/1958 (Peirce) and 1939 (Austin, the same source cited by Stroud and Franks).
It’s a documentable fact, however, that Van Til used “transcendental argument” in the contemporary sense prior to 1939. From his book A Survey of Christian Epistemology:
A truly transcendental argument takes any fact of experience which it wishes to investigate, and tries to determine what the presuppositions of such a fact must be, in order to make it what it is. . . . Now the only argument for an absolute God that holds water is a transcendental argument. . . . [T]he transcendental argument seeks to discover what sort of foundations the house of human knowledge must have, in order to be what it is.
Clearly Van Til is using the term “transcendental argument” in the modern sense, i.e., an argument that discloses the necessary preconditions (or presuppositions) of human knowledge and experience. Although SCE was published in 1969, it originated as a 1932 syllabus with the title “The Metaphysics of Apologetics.” It’s possible, I suppose, that Van Til added the references to “transcendental argument” sometime after 1932, but that’s unlikely for two reasons. First, the usage isn’t merely incidental. The entire four-page section entitled “Transcendental” in chapter 1 of SCE is devoted to discussing the need for a “transcendental argument” and “transcendental method” in Christian epistemology and apologetics. Secondly, Van Til had previously used the term “transcendental argument” seven times in his 1927 doctoral dissertation, “God and the Absolute.” A sampling:
Without the conception of a self-sufficient God our human experience would be meaningless. It is well to note at once the nature of the argument; it is transcendental and not formally logical. An argument for the existence of God based on formal logic would imply the ability to define God and arrive at a comprehensive rationality of all our experience. A transcendental argument on the contrary, is negative in so far that it reasons from the impossibility of the opposite.
But we have already indicated that the Beyond of Idealism is quite distinct from the Beyond of Theism. The latter is not the product of a priori thought and for that reason asks no destruction of spatio-temporal reality nor even any transmutation. On the contrary it is a concept obtained by transcendental argument and therefore necessary for the reality of time-experience.
Theism, accordingly, does not accept the challenge on this basis. It holds that anyone who seeks unassisted to cross the whole channel or perish will more than likely perish; scepticism is the only alternative to Theism. Theism too is sceptical as to man’s ability to know the whole of truth but it has sought by transcendental argument to establish the philosophical tenability of the existence of a timeless reality which is completed self-conscious actuality in which it rests for the validity of our knowledge as far as it goes.
It’s clear enough that Van Til is using “transcendental argument” in a sense that aligns, more or less, with contemporary usage.
So, does Van Til win the prize? Well, perhaps not. The potential spoiler is supplied by Christopher Hookway, also writing in the volume cited earlier:
In 1902 [C. S.] Peirce attempted to write a logic text, the Minute Logic (unfinished but excerpts appeared in his Collected Papers in 1932). The first chapter surveyed some approaches to logic including a discussion of the transcendental method of Kant and Hegel. It concluded ‘I have never met with an attempt to state a transcendental argument with precision which began to convince me’ . . . Peirce’s easy use of the phrase suggests that he saw nothing innovative in it.
Perhaps we must concede that Peirce deserves the credit for the first known use of the exact phrase in the now-conventional sense. (Van Til was only 6 or 7 years old when Peirce wrote that logic text!) Still, the 1902 reference looks like an outlier in the literature. Peirce doesn’t engage in an extended discussion of transcendental argumentation (using that terminology) in the way Van Til does in his earliest works, and it appears he doesn’t use that phrase again until his much later writings. So, we can at least say this: Van Til is one of the earliest 20th-century thinkers to speak of “transcendental arguments” in the contemporary sense. He is also one of the earliest 20th-century advocates of such arguments. It’s too bad that he isn’t given more credit for those distinctions.
Honoring the distinction between what Kant probably meant by “transcendental argument” and what thinkers in the 20th and early 21st centuries have often meant by that phrase, we can safely say that Cornelius Van Til did not coin the more recent usage.
Translators of the Critique of Pure Reason (such as F. Max Müller in 1881 or J. M. D. Meiklejohn in 1899) used the phrase “transcendental argument” in its earlier (strictly Kantian) sense, and commentators such as Antoine Claude Gabriel Jobert, in “Ideas, or Outlines of A New System of Philosophy,” London: Simpkin, Marshall, & Co., 1848, Chapter V (“Examination of transcendental arguments, on universal and necessary truths being *à priori* ideas–Remarks of Dr. J. S. Mill”), continued to understand the expression in that way:
“We will now examine upon what transcendental argument it has been maintained that universal and necessary truths, or ideas, cannot be acquired through the channel of our senses.” (Jobert, 53).
However, the book _The Foundations of Belief, being Notes Introductory to the Study of Theology_, London: Longmans, Green, and Co., 1895, by Arthur James Balfour, uses the expression “transcendental argument” a few times and seems transitional. Although it’s clear in context that Balfour understands himself to be interpreting Kant’s usage, Balfour’s explication tilts decidedly toward the more recent meaning.
Andrew Seth reviewed and engaged with Balfour’s tome, and Seth quite clearly takes Balfour to mean something like what Van Til would later mean:
“Such, without following them into details, are the important conclusions which Mr. Balfour reaches by the method of argument he follows…. [T]he philosophical student can hardly fail to remark the striking resemblance of Mr. Balfour’s mode of argument to the transcendental method of Kant…. It is the transcendental argument stated with a luminous simplicity…. But the argument itself is in substance identical with that which Kant patiently dug from the debris of rationalism, and built into a system so palpably artificial in its details and so cumbrously pedantic in its terminology that the philosophical world has been engaged ever since in quarrelling over it interpretation. When we penetrate beneath the portentous phrases to the comparatively humble truth which they labour to express, Kant’s ‘objective unity of apperception,’ as the supreme condition of the possibility of experience is simply the assertion that the idea of ‘a nature’ or a rational system is not a conclusion from particular facts, but is involved as a postulate or presupposition in there being an experience of facts at all. And when, at the close of his investigations, he emphasizes the adaptation of phenomena to our faculty of cognition as proof of a harmony between sense and understanding, that is to say, ultimately a harmony between the world and the mind; when he argues that this adaptation justifies us in treating reality as everywhere rationalisable, and therefore *as if* it were the product of a Supreme Reason; this, in more scholastic form, and with Kant’s well-known reservation as to the merely regulative character of the ideas of reason, is neither more nor less than the argument *from the mere fact that we know*.” Andrew Seth, “Mr Balfour and His Critics,” _The Contemporary Review_ (Vol 70, London: A Strahan, 1896), 153-178:
In Seth’s epitome of Balfour’s argument, two factors stand out as especially pertinent. First, that Seth feels that the “objective unity of apperception” (or “transcendental unity of apperception”, as it’s often rendered) is the (not merely syntactic or semantic but ontic, i.e., somehow underlying) precondition to “there being an experience of facts at all”–a robust expression of what’s at stake in the more recent usage. Second, that Seth takes for granted that philosophy students “can hardly fail” to see that Balfour’s move is strikingly similar and, in fact, “in substance identical” to Kant’s.
Seth gives the more recent definition but asserts its identity with the original definition. What are we to make of this apparent conflation of the newer and older senses of “transcendental argument”? Perhaps simply that the people engaged in migrating, broadening, and amplifying the expression didn’t perceive themselves to be making changes or additions; rather, they–or at any rate Seth and his set–understood themselves to be unpacking what Kant had left tacit.
We see evidence of this in Charles Douglas’s review of Andrew Seth’s book _Man’s Place in the Cosmos, and Other Essays_, Edinburgh and London: William Blackwood & Sons, 1897. Douglas illustrates Seth’s digestion of Balfour by selectively quoting the former on the latter in “Critical Notices”, _Mind: A Quarterly Review_ (Vol 7, Oxford UP, 1898), at page 100:
“If by premises… we understand either isolated intuitions or the particulars of sense, then it may be said that the transcendental argument neither starts from premises nor arrives at a conclusion. Yet, in a more vital sense, experience itself, as a concrete fact demanding explanation, constitutes the premise from which we advance (or rather regress) to its implied condition or explaining cause” (Seth, Man’s Place, 281).
The fact that Douglas chooses this pointedly post-Kantian illustration from Seth to represent the gist of Seth’s reading of Balfour suggests that Douglas understands that reading to be both uncontroversially Kantian and a fair take on Balfour. So it seems the meaning of “transcendental argument” was in unrecognized transition.
Since these occurrences of “transcendental argument” in the fresher sense date to the 1890s and Van Til was born in 1895, that question seems settled (unless we disagree that these usages are fresh enough to moot the issue). But (at least) two more interesting questions abide, for anyone seeking a thesis topic:
First, exactly which of Van Til’s advisors, courses of study, or opportunities for exposure put him in touch with this academic discussion, which was already underway? And, second, Balfour explicitly considers the Supreme Being as a fulfillment of the transcendental precondition, so… did Balfour get there first, well before Van Til, as far as TAG goes?
Fascinating! Thanks so much for writing this up.
It didn’t click when I first read your comment that Andrew Seth was later known as Andrew (Seth) Pringle-Pattison. VT interacted extensively with P-P in “God and the Absolute” and his other writings, and was clearly very familiar with his material. So perhaps that was one of VT’s main points of contact with the developing discussions of TAs.
Timothy McConnel (“The Influence of Idealism on the Apologetics of Cornelius Van Til”) notes that VT was very interested in Seth’s Hegelianism and Personality and The Idea of God in the Light of Recent Philosophy. Notably, the first of these was the published version of Seth’s “Balfour Philosophical Lectures.”
Wikipedia mentions that Seth adopted the surname extension to satisfy the conditions of a bequest. Amusing.
In any case, these facts dovetail nicely, and Van Til’s deep interest in Personalism offers a later juncture with _Hegelianism and Personality_, so everything tracks.
It’s probably worth noting, too, that Balfour made an evolutionary case against naturalism as early as 1914, arguing in his Gifford Lectures (published as “Theism and Humanism: Being the Gifford Lectures Delivered at the University of Glasgow, 1914,” Hodder and Stoughton: George H. Doran Company. 1915) that illusions or delusions (or itches or sneezes) might be just as conducive to survival or adaptation as reliably acquired sensory information is, such that strict naturalism needn’t favor truth-directed faculties. Sound familiar?
Indeed!
Victor Reppert mentions Balfour’s argument in his essay “The Argument from Reason” in The Blackwell Companion to Natural Theology: “The earliest post-Darwinian presentation of the Argument from Reason that I am familiar with…” He suggests it may have been part of the inspiration for Lewis’s anti-naturalism argument in Miracles.
Pingback: Did Cornelius Van Til Coin the Term “Transcendental Argument”? – Math Is Christian