Meanwhile I have had a few reactions to my first two blog postings, especially to the first one on Mickey Mouse linguistics. Let me reply to them in chronological order.
First Alex Jaker’s comment of March 4th with regard to the Mickey Mouse posting. Alex, who at the moment works as a field linguist in Alaska, proposes not to speak about “Mickey Mouse” but about “post-structural” linguistics, in analogy with “postmodern” movements in various areas of cultural life. I think Alex is right in drawing this parallel. I also think he is right in positing that this kind of approach to language is a weird mixture of quasi-positivist science (statistics) and, if I am reading him correctly, introspective methods letting linguistic gut feelings drive what passes for theory—an approach that puts bashing opponents above developing an academically respectable theory and has, therefore, never solved any real empirical problem. Even so, I prefer “Mickey Mouse linguistics”, as that term shows the caricature to which the input frequency fans reduce natural language, with all its ever new surprises and unexpected depths. It is also a term that stings, which is what it needs to do.
Then the comment of March 20th by my good friend Nigel Love, who calls my attention to John Taylor’s book The Mental Corpus (OUP 2012). Many thanks, Nigel, I have just got the book from our library and will discuss it soon in a later blog. It looks quite challenging. Nigel also commented on my examples (2a,b), I don’t think either Harry or John were late and I don’t think either Harry or John was late, respectively, as well as on my examples (3a,b), Only last summer he made headlines and Only last summer did he make headlines, respectively. I will come back to this latter pair of sentences soon, in a later blog, when I deal with Subject-Auxiliary inversion after fronted (semi)negative adverbials.
As regards the first pair of sentences, Nigel says that he can’t see any semantic difference between the two. This is always a difficult question in linguistics, as our data consist mainly of ‘protocol’ judgements on linguistic well-formedness or possible meanings. Of course, Nigel is a native speaker of English and I am not, though my command of English is pretty good, I dare say. Clearly, if Nigel is right for speakers of English generally, I must give up that point. But if there are a significant number of speakers of English who share my intuition, I do have a point. For the moment, all I can do is present some highly anecdotal evidence, consisting in unprimed judgements uttered by a few members of the institute where I work, all native speakers from diffrerent parts of the English-speaking world. They all agree that (2a), with the plural were, sounds better and is more natural than (2b), with was, both meaning “I think that neither Harry nor John was late”. When interpreted as “it is not so that I think that either Harry or John was late”, sentence (2b) gets more acceptable, but sentence (2a) cannot be interpreted that way. Provisional upshot: (2a) is both unambiguous and well-formed, and means “I think that neither Harry nor John was late”; (2b) is forced in that meaning, but gains in acceptability when taken to mean “it is not so that I think that either Harry or John was late”—a meaning that (2a) cannot have. So I am afraid I have at least a good chance of having a point, although I have to admit that to settle the issue it would be necessary to gather sufficient relevant data in a professional, experimental way—something we should perhaps do more often in linguistics.
I also had a comment by John Joseph, the author of the large book Saussure (OUP 2012). I admire both the book and its author enormously, yet I sounded a slightly critical note on his treatment of Hippolyte Taine in my second posting. John replied by email, not by way of a comment to the blog. Here is what he wrote to me:
“Thank you, Pieter, especially for the kind words about my book. I don’t think though that I’ve been inconsistent. Again, what I wrote is (I’m copying it from your blog): “The claim that Taine exerted a unique and universal influence on Saussure’s conception of language is far too strong.” [emphasis added]
It’s that specific claim, made by Aarsleff, that I think is far too strong. The absence of any reference to Taine anywhere in Saussure’s vast paper trail doesn’t prove anything in itself, nor do I claim that it does. On the other hand, for every one of the ideas which Aarsleff claims could only have come to Saussure via Taine, there are other sources whom we know, from documentation, that Saussure read or studied with. Hence I go on to write: “Similar claims could be made for dozens of nineteenth-century writers whose ideas show certain affinities to Saussure’s. What cannot be determined, in the absence of any documentation that Saussure read them, is whether there was ‘influence’ from them to him, or whether perhaps both they and he had drunk from the same trough — some common source, or ideas that were ‘in the air’ at the time.”
I would be perfectly delighted if a set of notes turned up which the young Saussure had made on his reading of Taine, in which he comments on specific ideas, comparable to the notes which he made on his reading of Victor Egger’s La parole intérieure. I was especially hopeful of finding something of the sort when I discovered that Taine had given a series of lectures at the University of Geneva when Saussure was a student there — but I found no indication that he attended them. In any case, having read the lectures, they deal with things very far from what Aarsleff thinks influenced Saussure.
My criticism of Aarsleff is based strictly on what I see as his jumping to a conclusion that is too firm for the evidence to sustain. I tried to make clear in my book that I don’t believe anything in Saussure’s conception of language lacks historical precedents, so it would take nothing away from him if Taine turned out to be the source of all that Aarsleff claimed.
Of course, the last thing I would want to do is antagonize John Joseph on illegitimate grounds, the more so because I am a great admirer of his book. Yet it would be useful if he could quote a few examples of “other sources whom we know, from documentation, that Saussure read or studied with” for the ideas that are central to Saussure’s thought and that Aarsleff claims came from Taine. For the rest, I have no problem giving John Joseph his due, as the question is probably more a matter of degree than of absolute opposition. The reason I raised the issue was twofold: I sense a general but in my view unjustified disparagement with regard to Aarsleff, and, unless proven otherwise, I am convinced that Taine did have an enormous influence on Saussure’s thought. The question is, at any rate, worth exploring in greater depth.
Then, another good old friend, Ad Foolen, of the Radboud University Nijmegen, pointed out, again by email, an article by Rong Chen, of California State University in San Bernardino CA, entitled ‘Subject auxiliary inversion and linguistic generalization: Evidence for functional cognitive motivation in language’ and published in Cognitive Linguistics 24.1 (2013), pp. 1–32. This article is relevant with regard to the examples (3a,b) in my Mickey Mouse blog, mentioned above in reference to Nigel Love’s comment. I will revert to this question, and to Chen’s paper, in a later blog, next week or so.
Finally, today I received an anonymous comment, by someone calling himself ‘Googlehupf’, who complains about what he (I assume it’s a he) sees as my proprietary use of the term “proper linguistics”, saying that “linguists are now interested in much more than grammar, and […] do not try to hide their being human when doing research in a rich variety of ways.” Although I am not happy with the fact that this writer prefers to stay anonymous, I am very happy to reply to him, since he naïvely and candidly demonstrates an attitude which the more prominent figures in cognitivist linguistics and related branches of the subject, such as input frequency linguistics, are usually much more reticent about. Roughly speaking, it is the attitude implying that one is, in some mysterious sense, more ‘human’ to the extent that one is less ‘formal’ in one’s academic work, and that this aversion to ‘formalism’ is or should be academically respectable.
This attitude is prominently found in the works of the mid-20th-century Catholic philosopher Jacques Maritain (1882–1973), especially with regard to logic. For Maritain, the new mathematical logic has ‘dehumanized’ the venerable traditional Aristotelian logic with its supposed links to metaphysics and the human soul. What Maritain, and our anonymous Googlehupf, have failed to see is that formalization is nothing but a method to formulate one’s scientific theory in a maximally precise way, no matter what the theory is about. Rejecting formalization as ‘dehumanizing’, Maritain and those sharing his confused ideas resort to mysticism and obscurantism, putting forward ideas about mysterious, often religiously inspired, entities like a totally undefined ‘soul’ or an unknown ‘human essence’, which, of course, have no place in science.
What scientists strive for, instead, is a rational explanation in terms of, preferably causal, factors yielding regular observable consequences that confirm or disconfirm the theory. If the theory is about the workings of human language in the human mind, then the scientist will strive for a maximally formalized theory of that. We know that a fully formalized analysis and description of language is not (yet) possible, as language is intimately connected with the rest of cognition and cognition as a whole is as yet far too big and too complex to be captured in formalized terms, though all sorts of cognitive ‘mechanisms’ have been identified and described in exact terms. Given this state of affairs, we see how far we can get. There is nothing ‘dehumanizing’ about that. On the contrary, the better we understand the complexities of the mind, and thus of language, the more we will appreciate the value of human nature.
The flight into obscurantism inevitably leads to absurdities. One absurdity, in the case of our friend Googlehupf, is that, apparently, the mechanical counting of frequencies is taken to be more ‘human’ than sorting out what system underlies observed linguistic phenomena. Another is that making the mind look more ‘flat’ by reducing linguistic and other knowledge to statistical frequency measures is taken to do more justice to ‘human’ values than investigating the depths and complexities of the mind, as is done in serious linguistics. I know, of course, that the Googlehupfs of this world are impervious to any amount of rational argument, but I also know that less fundamentalist minds, still open to uncluttered, rational reasoning, are all too often also still susceptible to the lure of ‘mystery’. It is to keep these away from the black caves of kabbalism and esotericism that I am writing this.
Googlehupf, moreover, claims the right to be “interested in much more than grammar.” I would be the last to deny him that right: the world would be a weird and awful place if one didn’t have that right. But not every ‘interest’ has a right to be pursued in an academic context, since universities are paid for with money provided by the community, one way or another, for the explicit purpose of enhancing reliable knowledge about the world, including ourselves. If an ‘interest’ fails to provide reliable knowledge, for example because it allows for the ignoring of crucial data or rejects sound ways of reasoning and theory-formation, then that ‘interest’ has no place in academia, no matter how many followers it has. This is why religions, or the likes of Googlehupf, have no place in our universities—though they are fully entitled to an entry in the phone book.