LINGUIST List 9.1298

Fri Sep 18 1998

Sum: Go+AND+Verb Constructions

Editor for this issue: Scott Fults <scottlinguistlist.org>

Directory

1. Anatol Stefanowitsch, Sum: GO+and+VERB constructions (long)

Several weeks ago, my colleague Stefan Gries <StThGriest-online.de> was so
kind as to post a query for me concerning (i) the semantics of the English
go+and+verb construction, (ii) corresponding constructions with other verbs
(such as run+and+verb, come+and+verb, try+and+verb, etc.), and (iii)
related constructions in other languages. I did not receive many replies
dealing with the first issue, but I did receive some dealing with the
second, and many detailed replies dealing with the third. Accordingly,
this summary consists of three parts: the first part is a sketch of
traditional descriptions of the sematics of the English go+and+verb
construction (in the hope that it might spawn some additional replies); the
second part summarizes the replies dealing with related English
constructions; the third part summarizes the replies dealing with the
related constructions in other languages. I received so much data that I
can only repeat the tiniest portion of it here, but I will make more
available to anyone who is interested.


1. The English go+and+verb construction

Dictionaries and descriptive grammars suggest that the go+and+verb
construction serves to express that the action described by the second verb
is "thoughtless, unfortunate, or silly" (Random House), "foolish,
unreasonable, or unlucky" (OED), or that it indicates "surprise or shock,
often showing disapproval" on the part of the speaker (Newbury House). In
addition, Tamara Al-Kasey suggested (in a discussion on LINGUIST in 1994,
Issue 5.1091.1) that it conveys a sense of "deliberateness." To confuse
matters further, a cursory glance at the Cobuild Bank of English reveals
that it is often used in making both friendly and unfriendly suggestions.
Finally, English coordinated verb constructions in general are sometimes
claimed to "belong[] to informal style" and in many cases to "have a
derogatory connotation" (Quirk et al.'s Comprehensive Grammar of English).
Apart from the obvious issue of bringing some order into these suggestions
and empirically validating the observations on which they
are--presumably--based, three questions strike me as particularly
interesting:
(i) Are the different meanings conveyed by the go+and+verb construction in
any way synchronically related? I would think so; they are probably the
result of relatively systematic (metaphorical or metonymical) extensions of
'go,' since similar arrays of meanings can also be found in other
'idiomatic' expressions with 'go' (cf. e.g. Radden 1996);
(ii) By what mechanism do the meanings of 'go' and the respective second
verb combine?
(iii) To what extent can the go+and+verb construction and the go+verb
construction be described as products of some kind of grammaticalization
process?
The literature I am aware does not deal with these questions at all, but
instead discusses mainly the formal properties of the construction, often
in relation to the go+verb construction. The relevant works are: Shopen
(1971), Carden and Pesetzky (1977), Pullum (1990), which contains a
critical evaluation of the two first-mentioned as well as some additional
sources, Jaeggli and Hyams (1993), and Cormack and Smith (1994).


2. Related constructions in English

I received two replies concerning related English constructions, both of
them dealing with the try+and+verb construction.

2.1 Fred Baube <fredrodan.moremagic.com> speculates that the construction
with 'go' served as a model for the one with 'try.'
The idea that one verb+and+verb construction served as a model for others
is quite interestenig. I am not aware of any work on this issue, but I have
checked the OED to see whether Fred's suggestion is at least compatible
with the historical data. Here is what I found: the first occurrence of
'try and' is given as 1686 (the example is: J. S[ergeant] Hist. Monast.
Convent 9 'They try and express their love to God by their thankfulness to
him'). The first occurrence of 'go and' is given as 1000, but the first
example where 'go' is clearly not used literally is 1755 (the example is:
H. Walpole Corr. cclxvii. (ed. 3) III. 105 'Don't go and imagine that L
1,200,000 was all Sunk in the gulph of Madame Pompadour'). Thus, It seems
possible that 'go and' served as a model for 'try and,' but the reality
probably much more complicated (as is usually th case).


2.2 Dawn Nordquist <nordquisunm.edu> has worked on the construction from
a discourse perspective, and reports that it is "used in contexts for
indicating that the complement verb is unlikely to be completed, or there
will be difficulty in completing it." These findings are to be published as
Nordquist (1998). Dawn also drew my attention to a paper dealing with the
difference between the try+and+verb and the try+to+verb construction (Lind
1983).


3. Related constructions in other languages

3.1 Polish

Lukasz Pielasa <llukekki.net.pl> replied with a discussion of the Polish
"wziac i zrobic" ('take and do') construction (due to the anglocentricity
of the ASCII code all diacritics are missing in these and all following
examples. I have stuck to the transliteration devices used by the authors
of the original messages). He gave the following examples:

(1) Jak go uderzylem, to on WZIAL I sie PRZEWROCIL
 When I hit him, then he took self fell
 'When I hit him, he simply fell down'

(2) Poprosila mnie, wiec WZIALEM I POSZEDLEM.
 She asked me, so I took and went
 'She asked me, so I just went there.'

Examples like these led to his initial speculation that "the meaning of the
Polish [construction] is 'just do'." I then mentioned Ekberg's analysis of
the same construction (Ekberg 1993a). She mentions the Polish construction,
but does not deal with it in detail. She analyzes the semantics of the
corresponding Swedish construction roughly as 'begin to VERB.'
In a second reply, Luke then analysed the meaning of the construction as
follows: "apart from the meaning 'just do,' the use of this construction
implies that the action was unexpected, sudden, unplanned, even contrary to
expectations." He gave many examples supporting this analysis, of which I
will just repeat two:

(3) Nikt w niego nie wierzyl, a on WZIAL I WYGRAL
 'Nobody believed in him, and he (suddenly) won/but he won'

(4) *Od lat planowal wyjazd do stolicy, i WZIAL I WYJECHAL
 *'He'd been planning to leave for the capital for years, and he left'

It seems to be possible to cancel the 'unexpectedness' reading (which makes
(4) unacceptable) in favor of the 'just do' reading already encountered in
(1) and (2), by changing 'a' ('and') to 'wiec' or 'to' (both meaning 'so'):

(5) Od lat planowal wyjazd do stolicy, wiec/to WZIAL I WYJECHAL
 He'd been planning to leave for the capital for years, so he finally
 left'

To sum up, Luke's account seems to be roughly compatible with Ekberg's, but
is much more fine-grained, indicating that more research is needed. With
respect to the English go+and+verb construction, it is interesting to note
that it is (like the take+and+verb construction in Polish) sometimes used
to express unexpectedness.
Finally, it should be mentioned that Irish English is also reported to have
a take+and+verb construction (cf. Ekberg 1993a). I have personally never
come across it, but here is one of the few possible candidates I found in
the Cobuild Bank of English:

(6) "Frounce felt sorry for the boy, so we give him the whole
 nine yards. She paid old weaselly Beasley over two hundred
 dollars for the legal papers, and look what's TOOK AND
 HAPPENED now." He swung his arm in a wide semicircle.
 "Dumbest goddamned thing I ever did."

The context is not really sufficient to determine the meaning of "take and
happen" here, but there seems to be a connotation of misfortune and
disapproval which would make the construction similar to the go+and+verb
construction.

3.2 Japanese

John Mackin <jmackinflm.se.fujitsu.co.jp> reports that "the Japanese
language makes common use of VERB+AND+VERB constructions," and describes
the syntax of these as <non-finite verb stem>+TE+<finite verb stem>, where
"normally <verb>+TE is a verbal conjunctive meaning 'and then' but certain
finite verb forms turn it into an idiom." Again, I will only repeat two
examples here: example (7) shows a fairly literal use, (8) shows a more
idiomatic use:

(7) Itte kaeru
 Go:CONJ come
 'Going but will come back/I will be back again today'

(8) Yatte shimatta
 Do:CONJ finish
 'I/he sure did it!/It's a done deed'

I am sure there must be work on these constructions, so if anyone knows of
relevant references, please come forward.


3.3 Scandinavian languages

3.3.1 Danish

Line Hove Mikkelsen <linecogsci.ed.ac.uk> and Soren Harder
<sharderling.hum.aau.dk> both drew my attention to constructions of the
following kind:

(9) GAA (rundt) OG TAENKE paa den eksamen hele tiden
 go (around) and think about that exam all time.the
 'be thinking about that exam all the time'

(10) SIDDE OG STUDERE hele tiden
 sit and study all time.the
 'be studying all the time'

(11) LIGGE OG LAESE i solen
 lie and read in sun.the
 'be reading (lying down) in the sun'

In these examples the verb+and+verb constructions are used "to convey
progressive aspect" (Line), i.e. they are "more or less equivalent to the
English gerundive, which does not have any direct morphological equivalent
in Danish" (Soren). The verbs 'gaa' ('go/walk'), 'sidde' ('sit'), and
'ligge' ('lie') apparently retain their lexical semantics to different
degrees: the latter two actually require that the subject of the sentence
be sitting/lying while carrying out the action denoted by the second verb.
"'Gaa' on the other hand can be used in these constructions without much
implication of actual (physical) walking movement" (Line):

(13) Jeg GIK OG SPEKULEREDE paa ...
 I walked and speculated about
 'I was thinking about ...'

(14) Jeg GAAR OG SPEKULERER paa ..
 I walk and speculate about ...
 'I'm thinking about ...'

Soren suggested the following characterization of the construction's
meaning: "'gaa' is used for states leading (slowly) into other states or
actions."
Danish also has the take+and+verb construction mentioned for Polish in 3.1
above, but since Lena Ekberg treats this construction in considerable
detail, I will not deal with it here, except for noting that, according to
Soren, it is used to indicate "a suggestion to do V." This is interesting,
because it is a use that is also found for the go+and+verb construction in
English.

3.3.2 Swedish

Jan K Lindstrom <jklindstrling.helsinki.fi> noted that the verb+and+verb
construction forms a large but "not necessarily unified expressive pattern
in Swedish." He gave the following examples for the gaa-och-V construction
(from a corpus search of literary texts):

(15) Hon ska gaa och dansa paa kva"llen
 'She will go and dance in the evening'

(16) Johan hade gaatt och gift sig
 'Johan had gone and got married'

(17) Att pojken skulle gaa och do" saa tidigt
 'That the boy should go and die so young'

(18) Naagot jag har gaatt och ta"nkt mycket paa
 Something I have gone and thought a lot about
 'Something I have been thinking about'

Jan notes that these examples "are not semantically unified": (15) involves
literally going somewhere, whereas (16) has a connotation of disapproval,
(17) has a connotation of misfortune. All three uses can also be found in
English, as the translations of the examples show. In addition, (18) is an
instance of the aspectual use already shown for Danish. Other verbs can
also be found in the verb+and+verb construction, with different degrees of
"idiomatic" meaning:

(19) Kan du komma och tra"ffa oss i kva"ll?
 'Can you come and see us in the evening?'

(20) Han har varit och beso"kt sin fo"rsta ka"rlek.
 He has been and visited his first love
 'He has been away visiting his first love'

(21) Jag kan ta och putsa silvret.
 I can take and clean the silver.
 'I can begin cleaning the silver'

(22) Hon sitter och la"ser.
 She sits and reads
 'She is reading'

Examples (19) and (20) are fairly straight forward, literal uses, while
(21) is the take-and-verb construction already mentioned, and (22) is a
further example of the aspectual use. As was the case for Danish, the
aspectual uses retain some degree of lexical meaning.

3.3.3 References concerning the Scandinavian constructions

I am aware of only three sources dealing with coordinated verb structures
in Scandindavian languages: Ekberg (1993a, b), and Wiklund (1996). I am
sure there must be more, especially regarding the "aspectual" uses. Any
suggestions are still welcome. I would also like to know whether other
Scandinavian languages or areally related non-Scandinavian languages have
similar constructions. Jan drew some parrallels to Finnish, which I would
like to hear more about).


Acknowledgements

I would like to thank everyone who replied. My special thanks goes to those
of you already mentioned above, who so generously shared your data with me.
I was stunned by the sheer mass of examples as well as the detailed
explanations you provided in the original replies as well as the subsequent
correspondence. I hope to meet you all one day!

Any further replies to the original query or this summary can be sent to my
NEW email address <anatolrice.edu>, and will certainly be appreciated.


References

CARDEN, Guy, David Pesetzky. 1977. 'Double-verb constructions, markedness,
and a fake coordination.' Chicago Linguistic Society 13: 82-92.
CORMACK, Annabel and Neil Smith. 1994. 'Serial Verbs.' UCL Working Papers
in Linguistics 6, University College London.
EKBERG, Lena. 1993a. 'The cognitive basis of the meaning and function of
cross-linguistic take and V.' In Jan Nuyts, Eric Pederson (eds),
Perspectives on Language and Conceptualization. Belgian Journal of
Linguistics 8, 21-42.
EKBERG, Lena. 1993b. 'Verbet 'ta' i metaforisk och grammatikaliserad
anva"ndning.' Spraak och stil 3.
JAEGGLI, Osvaldo, N. Hyams. 1993. 'On the independence and interdependence
of syntactic and morphological properties: English aspectual COME and GO.'
Natural Language and Linguistic Theory 11.2: 313-346.
LIND, Age. 1983. 'The variant forms try and/try to.' English Studies 64:
550-563.
NORDQUIST, Dawn. 1998. 'Try and: A discourse analysis.' Proceedings of the
First High Desert Linguistics Society Conference, April 3-4, 1998,
Albuquerque, NM.
PULLUM, Geoffrey K.1990.'Constraints on intransitive quasi-serial verb
constructions in modern colloquial English.' In Brian D. Joseph, Arnold M.
Zwicky (eds), When Verbs Collide: Papers from the Ohio State
Mini-Conference on Serial Verbs. Working Papers in Linguistics, Dept. of
Linguistics, Ohio State University, 218-239.
RADDEN, Gunter. 1996. 'Motion metaphorized: The case of coming and going.'
In Eugene H. Casad (ed), Cognitive Linguistics in the Redwoods: The
Expansion of a New Paradigm in Linguistics. Berlin and New York: Mouton de
Gruyter, 423-458.
SHOPEN, Timothy. 1971. 'Caught in the Act.' Chicago Linguistic Society 7:
254-263.
WIKLUND, Anna-Lena. 1996. 'Pseudocoordination is subordination.' In Oystein
Alexander Vangsnes (ed), The Role of Gender in (Mainland) Scandinavian
Linguistics. Working Papers in Scandinavian Syntax 58, Dept. of
Scandinavian Languages, University of Lund.


Author's address:

Anatol Stefanowitsch
Rice University
Dept. of Linguistics - MS 23
6100 Main Street
Houston, Texas 77005-1892
email: anatolrice.edu