LINGUIST List 9.846

Mon Jun 8 1998

Disc: Educating Eve (and Hypotaxis)

Editor for this issue: Andrew Carnie <>


  1. Geoffrey Sampson, Educating Eve

Message 1: Educating Eve

Date: Wed, 3 Jun 1998 13:55:31 +0100
From: Geoffrey Sampson <>
Subject: Educating Eve

I am surprised at how words are being put into my mouth in the course
of the discussion sparked off by Feargal Murphy's review of
my book EDUCATING EVE (posting 9.734). In posting 9.814,
"bwald" writes "Sampson (according to Murphy's account,
which seems credible to me) denied ... ". In posting 9.773, "Seth"
writes "In response to Feargal Murphy's review ..., [j]ust for the
record, it appears that in EE Sampson is using ... "; and so on. Surely
this comes perilously close to scholarship as garden-wall gossip
("My friend Edna said that her neighbour told her ... "). Am I 
old-fashioned in presupposing that scholars who criticize a book will have
read at least the passages they criticize? In this case the entire
book (which was published in Britain and in the USA in 1997) is not
very long.

"Just for the record", let me respond to some comments made about the
parataxis/hypotaxis issue, which seems to be the passage of EDUCATING
EVE that has aroused most interest (though it occupies less than
two pages of the book). In the first place, I did not claim
as an original finding of my own that the early history
of Proto-Indo-European languages, and of Hebrew, shows an absence of
hypotaxis; as I pointed out in EDUCATING EVE, I am not qualified to
make such a claim. Nor did I say that 19th-century historical linguists
were united in believing that this claim was true; they were not.
What I did say, quoting writings published between 1895 and 1982,
was that the majority consensus, among those who are qualified, seemed
to be that the claim was true; and I made the point concrete, for a
non-technical readership, by contrasting the purely paratactic Hebrew
grammar of the opening lines of the Book of Genesis with the heavily
hypotactic structure of a modern English translation.

According to Murphy, EDUCATING EVE "cites great linguists like Karl 
Brugmann, Hermann Paul and Eduard Hermann as support for the notion that 
languages go from a paratactic stage to a hypotactic stage". This is just 
wrong: the only mention of Hermann Paul in EDUCATING EVE was to point out 
that he _disagreed_ with this view (though he seems to have been in a

Again, Murphy writes "Hermann Jacobi's 1897 work 'Composition und
Nebensatz' [sic, Jacobi's title began with the word 'Compositum'] concluded
that 'Maori, like early stages of PIE may not have had contained [sic]
a relative particle' in subordination. ... So Jacobi is claiming
that PIE did have subordination ... ". (The internal quotation is misquoted
from a discussion of Jacobi by W.P. Lehmann, 1974: 60; Lehmann's
actual words were "like Maori, early stages of PIE may not have contained
a relative particle".) As a statement of what Lehmann implied that
Jacobi believed, this is again just wrong: higher on the same page
Lehmann had clearly stated that Jacobi "assum[ed] ... that subordinate
clauses arose from coordinate clauses", i.e. that early forms of
language lacked hypotaxis. 

If Murphy was looking for someone to cite as opposing this idea,
he would have done better to choose Lehmann himself;
on the previous page (p. 59), Lehmann describes Delbrueck's discussion
of the origin of IE relative clauses as "confused" because it assumed
"that this process was evolutionary and that it reflected the development
of man's ability to introduce hypotaxis or subordinate clauses in addition
to the earlier parataxis". On the other hand, Lehmann does not demur from
the general idea that the early historical record shows IE languages
increasing in grammatical complexity; for instance, on pp. 173-4 he
writes "The earliest poems of the Vedas are transparent in syntax ...
Such syntactic patterns ... lack the complexity of Classical Greek and
Latin, or even Homeric Greek". One way of resolving the tension between
these two passages in Lehmann's book would be to take him as agreeing that
grammar has grown more complex in the period of historical record, but
as objecting to Delbrueck for thinking that this was a process of
biological rather than cultural evolution. If so, that would agree with
the position defended in EDUCATING EVE.

The most telling passage in Murphy's review of EDUCATING EVE is his remark 
that "the book will be applauded by those who are already favourably 
disposed towards its conclusions and derided by those who are not". This 
well reflects the extent to which nativist linguistics -- belief in what 
Steven Pinker calls a "language instinct" -- has turned into a closed 
ideology, whose true believers cut themselves off from the possibility of 
meaningful debate with sceptics. A long passage in Murphy's review 
criticizes me for using written rather than spoken examples in refuting 
Noam Chomsky's claim about the extreme rarity in English of questions of 
the form "Is the man who is tall sad?" In fact the main counterexample I 
used was a line from a poem commonly recited aloud by schoolchildren, so 
its written/spoken status is unclear; but even if it were a purely written
quotation, it would have been a valid counterexample. The claim 
of Chomsky's which I was refuting was stated in very broad terms:
"you can easily live your whole life without ever producing a relevant
example ... you can go over a vast amount of data of experience without
ever finding such a case" (in Piattelli-Palmarini 1980: 114-15). The
life experience of most adult native speakers of English includes
written as well as spoken language, so either is a legitimate source of
counterexamples to this claim; but of course it is far easier to search
the written record than it is to search for cases from speech, the
great majority of which is never recorded. By declaring that written
evidence shall not count, Murphy is just protecting nativist ideology
from refutation by ruling out of court the most convenient source of
counterevidence. Empirical science should leave itself vulnerable to
possible refutation, not invent arbitrary rules of evidence aiming to 
render it immune to attack.

Murphy repeatedly criticizes EDUCATING EVE as superficial because the
nativist writings it takes issue with include books written for 
the educated general reader, such as Pinker's THE LANGUAGE INSTINCT. 
My declared strategy in EDUCATING EVE was to refute all arguments for
linguistic nativism found in those writings which have actually been
influential in spreading this point of view to a wide public.
I identified those writings as including relatively technical material
published by Chomsky up to about 1980, and various more popular books
published in the 1990s, of which Pinker's THE LANGUAGE INSTINCT is the
most notable. In refuting the nativist arguments contained in this
body of writing, I referred to the technical literature when I needed
to; often I did not need to, because the arguments were either
logically fallacious, or based on straightforwardly false premisses
(or both). A bad argument is a bad argument, whether couched in 
technical jargon or simple words.

If Murphy feels that this is an inappropriate strategy for refuting
the "language instinct" idea, does that mean that those arguments which
have actually succeeded in swaying educated public opinion are admitted
to be unsatisfactory, but that the inner coterie have other, better
arguments which they could have deployed publicly but did not? To say
that would be perhaps a severer criticism of the intellectual
integrity of the linguistic nativism movement than anything I say

The concept of a "language instinct" is at root a reasonable, logically
coherent hypothesis about how human beings succeed in mastering language;
it just happens to be wrong. I am rather confident that many or most
readers who are willing to approach my book with a reasonably open mind
will agree that it makes that case. But, as we say in England, there's
none so deaf as those that won't hear.


W.P. Lehmann (1974), _Proto-Indo-European Syntax_, University of Texas
M. Piattelli-Palmarini, ed. (1980), _Language and Learning: The Debate
	Between Jean Piaget and Noam Chomsky_, Routledge & Kegan Paul.

Geoffrey Sampson

School of Cognitive & Computing Sciences
University of Sussex
Falmer, Brighton BN1 9QH, GB

tel. +44 1273 678525
fax +44 1273 671320
Web site
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue