The Ontological Fallacy

The term ontological fallacy has great currency in social philosophy, where it is used to denote the mistake of assuming that because there is a term for something, like a social institution, that the object it denotes really exists. A similar couple of fallacies are Whitehead’s “Fallacy of Misplaced Concreteness”, in his Science and the Modern World, which he engaged against a certain kind of scientific realism, and Marcuse’s “Reification Fallacy” in One Dimensional Man.

Whitehead’s version is well known:

“There is an error; but it is merely the accidental error of mistaking the abstract for the concrete. It is an example of what I will call the ‘Fallacy of Misplaced Concreteness.’ This fallacy is the occasion of great confusion in philosophy. It is not necessary for the intellect to fall into the trap, though in this example there has been a very general tendency to do so.” Science and the Modern World, 1926, p74f

A little later he says

“Thereby, modern philosophy has been ruined. It has oscillated in a complex manner between three extremes. There are the dualists, who accept matter and mind as on an equal basis, and the two varieties of monists, those who put mind inside matter, and those who put matter inside mind. But this juggling with abstractions can never overcome the inherent confusion introduced by the ascription of misplaced concreteness to the scientific scheme of the seventeenth century.” p82. Cf. also 85

However, I wondered who had come up with this fallacy. It is not listed in the Stanford Encyclopedia of Philosophy, nor in the Internet Encyclopedia of Philosophy, nor in my battered old copy of Dagobert Rune’s Dictionary of Philosophy, but you find it used all over the place, as early as the 1920s, as if everybody knew what it meant. Sometimes it appears that people are referring to the Fregean attack on the Ontological Argument, but this seems to be a casual use of the phrase.

A bit of hunting and I found what I think is the origin: an essay by F. C. S. Schiller in 1905, but he doesn’t use the phrase. Instead he takes F. H. Bradley through his disciple Alfred E. Taylor, author of Elements of Metaphysics (1903), to task for making two correlated errors:

In its essence, this would seem to be a form of the ‘ontological’ argument whereby a claim of our thought is turned into a revelation about reality. But in addition there is a twofold fallacy, viz. (1) an equivocation in the word ‘truth,’ which is used both of the internal self-consistency of thought and of its ‘correspondence with reality,’ and (2) the unworkable view of truth as the correspondence of thought with reality. [p357]

Schiller is a pragmatist, so he rejects both claims. It is a widespread error in philosophy to think that because we have, say, a clear and distinct idea of something, there must be a something to go along with it. It’s a kind of Word Magick. The world doesn’t much concern itself with matching our linguistic categories – in fact the causal arrow is, or at least should be, entirely the other direction.

The ontological fallacy is best known from Anselm’s rather abstruse argument that a conception of God must include all perfections, including existence, and so God exists. But reality doesn’t follow what we can conceive (and it is far from obvious that we can conceive of a fully perfect being, in my opinion, but that’s a different fallacy). Words should be attempts to make the world make sense. But as I discovered when I encountered the Universal Language Project, such as my namesake Bishop John Wilkins and Leibniz, among others, thought would capture the essence of reality, it is a longstanding fallacy, and a persistent one.

Schiller, F. C. S. 1905. Empiricism and the Absolute. Mind 14 (55):348-370.

17 thoughts on “The Ontological Fallacy

  1. Wikipedia mentions “hypostatization” as another term for it.

    And it also mentions as a related fallacy the “pathetic fallacy” (which, of course, is not a fallacy, but a metaphor) or the “anamistic fallacy”.

    I like to think of a triplet of “fallacies”:
    1) Seeing a pattern where none exists
    2) Objectifying that pattern into an object
    3) Claiming purpose directing that object

    And I think that this triplet is behind conspiracy theories, space aliens as the pilots of UFOs, and, of course, creationism.

    Not to mention my explanation of a triplet of fallacies.

    1. A related fallacy is the “Epistemic Fallacy”, which holds that if something exists it can be known to exist.

      The Triplet Fallacy? Don’t get me started. I can think of three things that are wrong with it…

  2. Thats extremly intresting. When I started studying this subject seriously my understanding of philosophy was terrible (it’s still far from perfect)

    I identified James Burnett as an interesting possibility as some of the narrative he uses with regard to the oragutang is older and associated with the wild-man. Fact Burnet was a philosopher was irrelevant. But on reading him
    I could see that he had in a way the same interests as I do, but the work was very disordered and non-empirical. Bit like a distant mad elderly relative. Although thats slightly unfair as on one or to matters he was rather good.

    In order to deal with it further and in an attempt to understand more of the history of this period
    I went to the book shop and rifled the philosophy shelf to see if I could find what seemed most ‘on topic’.

    Foucault’s Archeology of Knowledge seemed to be attempting to discuss the history surrounding the subject. I knew nothing with regard to post modern truth claims and at first despite Foucaults use of language (drives me nuts) his interest in speech performance theory, his interests in history are very similar to my own.

    But it was when I hit the chapter on Bishop Wilkins that it all fell apart. I think there are some problems with the way taxonomy is used in folklore but it has been central to establishing the subject on a clear empirical basis. I would have nothing to study without it.

    One of the things that helped me greatly was the other Wilkins interest in barnacle geese. I had been researching the subject for some time as it has been used by all social groups ansd I also have a very good idea of how it has been used in folklore. I could see it had some relationship with species as well but that is as far as I had got.

    Mr Wilkins article on the barnacle goose is one of the best I had read. I got in touch, he set me straight on a few problems I had been having.

    Then i remebered some of the papers I had read
    when Tancred Robertson a close associate of John Ray and Edward Lluwyd realy strips the language that surrounds the goose and presents it’s anatomy in clear, simple terms.

    It was an important moment for me in grasping what science was. As my meeting with the first Wilkins had been in rejecting post- modernisim.

    I had been looking carefully at Sir Thomas Brown and how he was analysing and rejecting but it was the goose and species that really hit home the work and effort applied at this period.

    But I do have a problem, when I speak about inflection range. I checked afterwards, it looks almost identical to Deluze. Yet the conclusions are very diffrent. I am sure it is in an understanding of inflection and it’s range that the diffrence must lie. So to use the language of Faucault part of what I have to do is to work out where the break occurs in his line of thought, the moment where his thoughts turn in a very diffrent direction from my own.

    Not a pleasant task. But I suspect this may help.
    Without identifying clearly where the diffrences lie and what I am doing. It looks like a bad post modern joke.

    Which it certianly is not. Given my interest in word play I have avoided any mention of the the double ‘Wilkinson blade’ effect.

    Cuts through itchy and annoying stubble like a razor. I find.

  3. “The tendency has always been strong to believe that whatever receives a name must be an entity or being, having an independent existence of its own; and if no real entity answering to the name could be found, men did not for that reason suppose that none existed, but imagined that it was something peculiarly abstruse and mysterious, too high to be an object of sense. The meaning of all general, and especially of all abstract terms, became in this way enveloped in a mystical haze…”
    A note by John Stuart Mill in a book by James Mill (Analysis of the Phenomena of the Human Mind (1869), Ch 14, “Some names which require a particular explanation”, Section 1, “Names of names”)
    (A nice summarization of the reification problem.)

    1. Thanks, on first glance I think that may answer a question. Most near I could come to seeing how a monster is captured in the field and investigated was Monboddo’s and other’s examination of French wild creature and one case in India.

      The initial enquires involve all social classes, each agree that a creature of a certian type has been caught. But in the naming of things and the narrative that surrounds the creature, there is much disagrement. The object being described is a diffrent thing. Yet no one see’s any contradiction in that fact (Monboddo would appear the exeption to that rule). A new narrative is constructed and the fact that all groups see it clearly appears as an important element in the construction.

      This form of evidence gathering by unreliable eye witness accounts was of course rejected by science, astronomy is one example where it became problematic and faults became clear.

      But certainly this form of constructing information and naming things was and still is a potent force.

  4. One other figure before Anselm is one of the last users of Hisperic Latin, Eriugena (Irish born).

    Ive yet to read Periphyseon,De divisione naturae
    His best know contribution but it does look interesting.

  5. Sorry to go on but I suspect this post has been the most fruitfull you have written from a personal perspective. As it should help me resolve some particular headaches, Ive identified but was unsure of where to look.

  6. “The word was made flesh and dwelt amongst us”

    John i. 14

    One from some old notes. But I can’t say I found it very inspiring or based my methods on it.

      1. Gnosticism in any meaningful sense of the term postdates Johannine theology; the actual direction of influence, where there is any, appears to go in the opposite direction. Much more plausible as a source is Middle Platonism, particularly as filtered through Hellenistic Judaism; Philo of Alexandria is the closest predecessor of which we have any extensive knowledge, and may actually be the immediate predecessor. That’s a nicer and more rational pedigree; but, of course, if there are any positions that commit an ‘ontological fallacy’, Platonistic positions are the least crazy ones to do so.

        For my part, I’m inclined to think that the label ‘ontological fallacy’, like a lot of ersatz fallacies, looks itself suspiciously like the sort of thing we get if we mistake thought for reality; as if our being able to label something a fallacy made it so. If (for instance) one were to say that Platonism is false because it commits the ontological fallacy, it would be no more informative than saying that Platonism is false because one of its main contentions about the relation between the abstract and the concrete is false: it’s not an explanation of error, as a real fallacy is, but simply a slightly more specific statement that it is an error, dressed up to look like an explanation for why it is an error.

      2. I think there were gnostic sects well before Christianity. Certainly Christian Gnosticism postdates Christianity for the obvious reason, but back when I did Johannine studies, most commentators thought that John’s gospel was written in reaction to various gnostic sects and religions.

        And yes, I think that Plato’s entire metaphysics is based on word games of this kind. That is, I think Platonism is a fallacious inference from words. The error is to think the fallacy means something.

      3. That indeed used to be a common view; it has become less popular because it made ‘Gnosticism’ such an amorphous term — really it just ended up meaning ‘paganism’ when talking about non-Christians and ‘Gnostics’ when talking about Christians (in a broad sense of the term). There is very little scholarly consensus about what counts as ‘Gnosticism’ in the first place; there are some Christian cases that are definitely in, and everything else that gets put in is put in by analogy, which is an unruly method of classification. Pre-Christian Gnosticism was never based on evidence of actual sects but was put forward as a hypothesis (and a rather vague one at that, given the lack of agreed meaning of what a pre-Christian Gnosticism would be) by Bultmann and others in order to explain the historical development of things like Johannine theology. It has been known for about twenty years now, though, that no evidence has been found to support that hypothesis — it makes more sense simply to speak of influences on Gnosticism, and divide them into the two very different groups of popular paganism and Middle Platonism.

        I’m not at all clear what your grounds for the second paragraph are. Platonism is obviously not a fallacious inference from words because none of the standardly Platonistic features — Forms and the like — are reached by arguments that begin with theses about words. The Forms are used to explain certain features of language, but this is a different thing. One can also argue that these arguments involve fallacies which are made easier because of the words in which they are described, but this is also a different thing entirely.

        And this is precisely one of the reasons why I’m tempted to propose as a joke a ‘fallacy of misplaced fallacies’ and put the ‘ontological fallacy’ under it — it does no work and explains almost nothing, and what it does explain, it explains misleadingly. Saying that something commits ‘the ontological fallacy’ is just to say it is wrong about how it distinguishes abstract from concrete; in this sort of context that just amounts to saying that certain key assumptions of Platonism are wrong. That’s certainly true, but that doesn’t mean any fallacy has been committed; fallacies require specific forms of misreasoning. There is nothing corresponding to this term ‘ontological fallacy’ that is identifiably a fallacy; unlike genuine fallacy-identifications it doesn’t clarify how reasoning itself works but merely rules out a wide variety of philosophical positions without explanation. The ‘naturalistic fallacy’ suffers the same problem: it’s not a fallacy, it’s a position someone dislikes for some reason, and the description for it is sophistically dressed up as if it were an error of reasoning rather than (at most) an error in assumption. Talking about the ‘naturalistic fallacy’ confuses assumed premises and inference-structures; and so does talking about the ‘ontological fallacy’.

      4. It seems to me that if there is an ontological fallacy, it implies an uncomfortable intellectual doppleganger. How are we to infer or abduct entities if not through concepts of those entities? To say our concepts should fit the world implies some impossible triangulation of reality by other means.

        All metaphysics is “word magic.” (And nothing expressed in language is truly “concrete”). The Fallacy of Misplaced Concreteness correctly applied pertains to the set of all things we can describe.

        I think you want to make the same claim as Leibniz in the end, only with a superior (more rarified) set of thoughts which “capture the essence of reality.” It seems (superficially, perhaps) its not his project you object to, but his overconfidence.

  7. Athanasius of Alexandria, ‘hoi anthropi’, the original soul of men. I am sure the Celtic legends of the wildman are Gnostic, it certainly seems to play on very similar themes.

Leave a Reply