There is an extensive literature
dealing with English imperative sentences. As is well known, these sentences
have no overt grammatical subject:
(1) Close the door.
There is general agreement among
scholars that these sentences have deep structures involving an underlying
subject you which is deleted by a transformation.
There is a widespread misconception
that utterances such as:
(2) Fuck you.
which also appear to have the
form of a transitive verb followed by a noun phrase and preceded by no
overt subject, are also imperative. This paper will study the syntax of
sentences such as (2). While it will offer only a tentative conjecture
as to what the deep structure of sentences such as (2) is, it will at least
demonstrate conclusively that they are not imperatives.
One characteristic of sentences
such as (2), which--as has been often noted--is an anomaly if they are
analyzed as imperatives, is the absence of reflexivization; while
(3) *Assert you.
is ungrammatical, (2) is not.
There are many other anomalies which are not so widely recognized. While
there are a large number of structures in which imperatives appear either
embedded in a matrix or with various adjuncts:
(4) I said to close the door.
(5) Don't close the door.
(6) Do close the door.
(7) Please close the door.
(8) Close the door, won't you?
(9) Go close the door.
(10) Close the door or I'll
take away your teddy-bear.
(11) Close the door and I'll
give you a dollar.
there are no such sentences
corresponding to (2):
(12) *I said to fuck you.
(13) *Don't fuck you.
(14) *Do fuck you.
(15) *Please fuck you.
(16) *Fuck you, won't you?
(17) *Go fuck you.
(18) *Fuck you or I'll take
away your teddy-bear.
(19) *Fuck you and I'll give
you a dollar.
Further, while ordinary imperatives
can be conjoined with each other, they cannot be conjoined with (2):
(20) Wash the dishes and sweep
the floor.
(21) *Wash the dishes and fuck
you.
(22) *Fuck you and wash the
dishes.
Similarly, sentences such as
(20) can be reduced to sentences with a conjoined verb if the two conjuncts
differ only in the verb; however, the fuck of (2) may not appear
in such a construction:
(23) Clean and press these pants.
(24) *Describe and fuck communism.
Likewise, there are sentences
containing the word fuck which are ambiguous between a meaning parallel
to (1) and a meaning parallel to (2):
(25) Fuck Lyndon Johnson.
This sentence can be interpreted
either as an admonition to copulate with Lyndon Johnson or as an epithet
indicating disapproval of that individual but conveying no instruction
to engage in sexual relations with him. When sentences with the embeddings
and adjuncts of (4) to (11) and (20) are formed, the resulting sentences
allow only the former of these readings:
(12a) I said to fuck Lyndon
Johnson.
(13a) Don't fuck Lyndon Johnson.
(14a) Do fuck Lyndon Johnson.
(15a) Please fuck Lyndon Johnson.
(16a) Fuck Lyndon Johnson, won't
you?
(17a) Go fuck Lyndon Johnson.
(18a) Fuck Lyndon Johnson or
I'll take away your teddy-bear.
(19a) Fuck Lyndon Johnson and
I'll give you a dollar.
(20a) Fuck Lyndon Johnson and
wash the dishes.
Consideration of these examples
makes it fairly clear that the fuck of (12a)-(20a) (henceforth fuck1)
and the fuck of (2) (henceforth fuck2) are two
distinct homophonous lexical items. These two lexical items have totally
different selectional restrictions, as is shown by the examples:
(26) Fuck these irregular verbs.
(27) *John fucked these irregular
verbs.
(28) Fuck communism.
(29) *John fucked communism.
Moreover, fuck2
has a peculiar restriction on the determiner of the following noun phrase,
a restriction not shared by fuck1, namely that the determiner
must be either definite or generic:
(30) Fuck these seven irregular
verbs.
(31) Fuck irregular verbs.
(32) Fuck all irregular verbs.
(33) *Fuck seven irregular verbs.
(34) *Fuck any irregular verb.
but
(35) Fuck seven old ladies by
midnight or I'll take away your teddy-bear.
(36) Fuck any old lady you see.
(the latter two involving fuck1).
It should be noted that the word "generic" must be interpreted in a sense
such that all is generic (cf. example (32)) but each is not:
(37) *Fuck each irregular verb.
Indeed, substitution into the
frame "Fuck____irregular verb(s)" is an excellent diagnostic test for genericness.
As example (35) makes clear, the two fucks also differ in their
potential for co-occurring with adverbial elements; while (35) is normal,
(38) *Fuck you by midnight.
is not. Moreover, note the examples:
(39) Fuck my sister tomorrow
afternoon.
(40) *Fuck these irregular verbs
tomorrow afternoon.
(41) Fuck my sister on the sofa.
(42) *Fuck communism on the
sofa.
(43) Fuck my sister carefully.
(44) *Fuck complex symbols carefully.
Evidently, fuck2does not allow any adverbial elements at all. This restriction suggests that fuck2 not only is distinct from fuck1 but indeed is not even a verb. Chomsky observes that the adverbial elements of (39)-(42) are outside of the verb phrase and that only elements within the verb phrase play a role in strict subcategorization of verbs. That principle would clearly be violated if fuck2 were a verb. While the "principle of strictly local subcategorization" proposed by Chomsky is in fact not valid in precisely that form, the fact remains that no case has been reported of any English morpheme which is unambiguously a verb and which allows no adverbial elements whatever. Since the only reason which has ever been proposed for analyzing fuck2 as a verb is its appearance in a construction (that of (2)) which superficially resembles an imperative but in fact is not, one must conclude that there is in fact not a scrap of evidence in favor of assigning fuck2 to the class "verb", and indeed, assigning it to that class would force the recognition of an anomalous subclass of verbs which violate otherwise completely valid generalizations about "verbs".
If fuck2 is
not a verb, then what is it? To make some headway towards answering this
question, let us consider the following expressions, which have much in
common with (2):
(45) Damn Lyndon Johnson.
(46) Shit on Lyndon Johnson.
(47) To hell with Lyndon Johnson.
(48) Hooray for Christine Keeler.
These expressions likewise exclude
adverbial elements and require the following noun phrase to be definite
or generic:
(49) Damn those irregular verbs.
(50) *Damn those irregular verbs
tomorrow.
(51) *Damn seven irregular verbs.
(52) Shit on all irregular verbs.
(53) *Shit on each irregular
verb.
(54) *Hooray for an irregular
verb last night.
Only rarely have hypotheses been
advanced as to the deep structure of expressions such as (45)-(48). One
hypotheses has been that (45) has an underlying subject God, which
is deleted. However, this proposal is untenable since it would exclude
the completely acceptable sentence
(55) Damn God.
and imply the grammaticality
of the non-sentence
(56) *Damn Himself.
It is interesting that in this
respect goddam works exactly like damn:
(57) Goddam God.
(58) *Goddam Himself.
While the assumption of a deleted
subject, God, has semantic plausibility in the case of sentences
such as (46) and (2), such an analysis must be rejected for the same reason
as in the case of damn, namely, the grammaticality of
(59) Fuck God.
(60) Shit on God.
and the ungrammaticality of
(61) *Fuck Himself.
(62) *Shit on Himself.
Consider now the semantics of fuck2, damn, to hell with, shit on, hooray for, etc. A sentence consisting of one of these items plus a noun-phrase has neither declarative nor interrogative nor imperative meaning; one can neither deny nor answer nor comply with such an utterance. These utterances simply express a favorable or unfavorable attitude on the part of the speaker towards the thing or things denoted by the noun-phrase. The fact that they have such a semantic interpretation explains the restriction on the determiner of the noun-phrase; the noun-phrase must specify a thing or class of things in order for the utterance to be semantically interpretable.
Note further the possibility
of using most of the words in question without any following noun-phrase:
(64) Fuck!
(65) Damn!
(66) Shit!
(67) Hooray!
These sentences indicate the attitude in question but do not specify what object that attitude is directed towards by the speaker.
The fact that sentences of the
form fuck2 plus NP are not known to be validly analyzable
as NP + VP in deep structure, the fact that they are not embeddable in
any sentences, and the fact that they allow none of the adjuncts which
all other sentences allow, makes highly plausible the hypothesis that they
should not even be analyzed as sentences--that the category "utterance"
be divided into two subcategories, "sentence" and "epithet" (the latter
class including utterances such as (2), (46) and (64)), that only "sentence"
and not "epithet" be embeddable within an utterance, that "epithet" involve
a lexical category of "quasi-verbs" (this category consisting of fuck2,
shit
on, etc.), that there be a phrase-structure rule
Epithet --> Quasi-verb NP
and that "Quasi-verb" appear
in no other phrase-structure rule.
In closing, I should mention
certain problems which I have not dealt with and which the reader should
be aware of. First, there is the matter of stress in "epithets". I know
of no non-ad-hoc treatment of the stress difference between
(78) Fuck you.
(79) Damn you.
Moreover, quasi-verbs have a
tendency to take primary stress. Stress may disambiguate (63) (although
the distinction is lost when contrastive stress is placed on the NP):
(80) Shit on the carpet. (=
Fuck2 the carpet.)
(81) Shit on the carpet. (=
Defecate on the carpet.)
A second matter which deserves a full treatment is the process of historical change whereby normal lexical items become quasi-verbs. I conjecture that fuck2 arose historically from fuck1, although the paucity of citations of fuck makes the philological validation of this conjecture difficult. However, it is clearly no accident that many quasi-verbs are homophonous with normal morphemes.
South Hanoi Institute of Technology
Revised version, Feb. 5, 1967
Message 2: Haj = Quang, not
Date: Sat, 2 Oct 93 10:04:55 -04Haj = Quang, not
From: Randy Allen Harris <raha@watarts.uwaterloo.ca>
Subject: Haj = Quang, not
A recent posting on the etymology of "Haj" included the following:
>... in a paper under
the name of Quang Phuc Dong
>of the South Hanoi
Institute of Technology: one of Ross's
>pseudonyms for his
essays in scato- and pornolinguistics. ...
(LINGUIST 4.777)
"Quang Phuc Dong" is
a nom de guerre (linguistique) not of Ross
but of James D. McCawley,
who "created the interdisciplinary field
of pornolinguistics
and scatolinguistics virtually on his own" (Zwicky, viii).
Ross, under the name
"E. Clifton Gamahuche", took the first and only steps
towards developing metapornolinguistics,
with his "Conjunctive Ordering"
(where, among other
observations, he notes that in the absence of Copula
agreement, the only
option is Reflexivization).
Quang's, Gamahuche's,
and similarly minded people's research can be found
in the recently reissued
_Studies out in left field: Defamatory essays
presented to James D.
McCawley on his 33rd or 34th birthday_, introduced
by Arnold Zwicky (whence
the above quotation), and edited by him et al.
The reissue is by John
Benjamins (Amsterdam and Philadelphia, 1992),
for about $15 U.S.,
and, like the original (1971), is not for the squeamish.
Randy Allen Harris
raha@watarts.uwaterloo.ca
Rhetoric and Professional
Writing 519
885-1211, x5362
English, U of Waterloo
FAX: 519 884-8995
Waterloo ON, CANADA,
N2L 3G1
DISCLAIMER: I am not the author of ENGLISH SENTENCES WITHOUT OVERT GRAMMATICAL SUBJECTS. The article clearly states that Quang Phuc Dong is the author. However, you can go to THIS SITE to see that, in fact, one James D. McCawley appears to be the author. For more on this subject, see set of postings from 1993: Disc: Last Word on Legends
Here's another article of interest by James D.
McCawley:
Dates In The
Month Of May That Are Of Interest To Linguists
Sadly, James D. McCawley passed away in 1999.
Note
from the director of the Linguistic Institute at the University of Chicago