������������� ���

�� �������������������������� ���������

John Newman john.newman@ualberta.caSally Rice sally.rice@ualberta.ca

University of Alberta, Canada

��� �

� V and V combination attracts little attention in narrowly syntactic approaches to English grammar.

����� ������� �

� V and V is not discussed as such.

� It would be a special case of Conjunction Reduction which applies to many kinds of conjoined clauses:

Robin drove out of Phoenix this morning and pro will arrive in Atlanta tomorrow.

Robert Van Valin and Randy LaPolla (1997). Syntax: Structure, meaning and function.Cambridge: Cambridge University Press. Pp. 520-523

���� ���������� ��

� Syntax of coordination is considered along with the lexical items that occur in coordination.

� “Asymmetric Constructions” = coordination where V1and V2 cannot be reversed without a change of meaning:

I fell off the ladder and broke my leg.

Rodney Huddleston and Geoffrey K. Pullum (2002). The Cambridge Grammar of the English Language. Cambridge: Cambridge University Press. pp. 1299-1305.

�������������� ��

� Symmetric at the abstract syntactic level (a la Van Valin and LaPolla), but asymmetric at the lexical/semantic level (a laHuddleston and Pullum, Givon etc.).

� A possible locus of grammaticalizationi-pile-la-be i-eno [Manam; Lichtenberk 2002]3sg.realis-speak-limiter-and 3sg.realis-lie‘He kept talking.’

����������������

� What kinds of verbs occur in the V and Vconstruction?

� Are there special meanings/ grammaticalizations which arise in the V and V construction?

� How are the two verbs in the V and V construction integrated semantically?

����������������

� What kinds of verbs occur in the V and Vconstruction?

� Are there special meanings/ grammaticalizations which arise in the V and V construction?

� How are the two verbs in the V and V construction integrated semantically?

� �����!�������������������

� Wellington Corpus� New Zealand English� 1 million words written (WC)� 1 million words transcribed spoken (WSC)

� British National Corpus� British English� 90 million words written� 10 million words transcribed spoken

�� ����"��!��

� Large enough corpus to see high frequency patterns

� Small enough corpus to make an exhaustive study of V and V

���� � ������� ����"��!��

� We identified all and occurrences and then manually identified sequences of V and V

sing and dance

� We allowed verb particlescome in and wait

� We excluded the copula and auxiliarieswas and will be

CLAWS Tagset

come and goneVVN (past participle)

coming and goingVVG (verbal -ing)

went and waitedVVD (Simple Past)

It comes and goesVVZ (3Sg Pres)

Wait and see! They go and wait

VVB (‘base’)

will wait and see,wants to wait and see

VVI (infinitive)

CLAWS Tagset

VVN (past participle)

VVG (-ing)

VVD (Simple Past)

VVZ (3Sg Pres)

VVB (‘base’)

VVI (infinitive)

SpokenWritten

V and V in the Wellington Corpus

98271VVN (past participle)

148248VVG (-ing)

267325VVD (Simple Past)

6490VVZ (3Sg Pres)

208113VVB (‘base’)

692328VVI (infinitive)

SpokenWritten

����������� �����

Number of lemma types in V1 and V2 construction

370

275

658

587

712828

0

100

200

300

400

500

600

700

800

900

V1ALL

V2ALL

V1WC

V2WC

V1WSC

V2WSC

Most frequent V1 lemmas in written WC

others, 1074 smile, 15

say, 15read, 15

sit, 16

try, 18

stand, 18

stop, 20

come, 80

go, 54get, 26 turn, 24

Most frequent V1 lemmas in spoken WSC

go, 462

come, 229try, 168

turn, 26 sit, 28

walk, 13stop, 13

look, 14

stand, 15wait, 17

ring, 21get, 36

others, 433

�# ���$%���&%���'���� ����"��!��

COME

GO

V1 of VVB in WSC

98

75

35

V1 of VVI in WC

279

3118

V1 of VVI in WSC

331256

105

V1 of VVD in WC

285

1030

V1 of VVD in WSC

125

87

55

V1 of VVB in WC

87

818

�# �(���$��� ��$�$)"

BNC samples based on 20,000 random and.

No verb particles after V1

Conversation (BNC) based on all (6,916) hits for and in Conversation sub-corpus of 4.2 million words

COME

GO

wBNC sample

35 4

13

sBNC sample

47

56

22

BNC Conversation

16

41

14

Past Tense V1 in Wellington Spoken Corpus

3.05attraction2161left10

4.80attraction3174rang9

8.09attraction481talked8

8.62attraction460stopped7

9.60attraction6290looked6

14.21attraction651got_up5

15.60attraction631stood4

25.73attraction1191turned3

117.38attraction55683came2

174.75attraction871509went1

CollostructionalStrength

RelationFreq. in V and V

Freq. in Corpus

Collostructional Strength: -log (Fisher exact, 10), the higher the stronger. Coll.strength>3 => p<0.001. Gries, Stefan Th. 2004. Coll.analysis 3. A program for R for Windows 2.x.

Past Tense V1 in Wellington Spoken Corpus

3.05attraction2161left10

4.80attraction3174rang9

8.09attraction481talked8

8.62attraction460stopped7

9.60attraction6290looked6

14.21attraction651got_up5

15.60attraction631stood4

25.73attraction1191turned3

117.38attraction55683came2

174.75attraction871509went1

CollostructionalStrength

RelationFreq. in V and V

Freq. in Corpus

Collostructional Strength: -log (Fisher exact, 10), the higher the stronger. Coll.strength>3 => p<0.001. Gries, Stefan Th. 2004. Coll.analysis 3. A program for R for Windows 2.x.

Past Tense V1 in Wellington Spoken Corpus

3.05attraction2161left10

4.80attraction3174rang9

8.09attraction481talked8

8.62attraction460stopped7

9.60attraction6290looked6

14.21attraction651got_up5

15.60attraction631stood4

25.73attraction1191turned3

117.38attraction55683came2

174.75attraction871509went1

CollostructionalStrength

RelationFreq. in V and V

Freq. in Corpus

Collostructional Strength: -log (Fisher exact, 10), the higher the stronger. Coll.strength>3 => p<0.001. Gries, Stefan Th. 2004. Coll.analysis 3. A program for R for Windows 2.x.

Past Tense V1 in Wellington Written Corpus

7.06attraction5319saw10

7.54attraction91998said9

10.80attraction6150sat8

12.83attraction670stopped7

13.35attraction8256looked6

14.79attraction10478went5

17.80attraction9140stood4

25.96attraction921got_up3

26.35attraction13173turned2

55.52attraction30563came1

CollostructionalStrength

RelationFreq. in V and V

Freq. in Corpus

Collostructional Strength: -log (Fisher exact, 10), the higher the stronger. Coll.strength>3 => p<0.001. Gries, Stefan Th. 2004. Coll.analysis 3. A program for R for Windows 2.x.

�����*����(���������������������� �

����������

� COME and GO together account for more than half of the tokens in V1 position in the infinitival, base, and past tense forms in the Wellington Corpus

� GO > COME as V1 in the spoken corpus� COME > GO as V1 in the written corpus� V1 position has fewer types of verb than V2

position

"+,�����-+��.� �� /��!��0

‘come’ and ‘go’ are the most common verbs in serial verb constructions in Kalam.

Pawley, A. and J. Lane (1998). From event sequence to grammar: Serial verb constructions in Kalam. In A. Siewierska and J. J. Song (Eds.), Case, typology, and grammar: In honour of Barry J. Blake, 201-227. Amsterdam and Philadelphia: John Benjamins.

"+,�����-+��.���1/�����%������� �0

‘come’ and ‘go’ are the only verbs which enter into a serial verb construction with other verbs.

Foley, W. A. and M. Olson (1985). Clausehood and verb serialization. In J. Nichols and A. C. Woodbury (Eds.), Grammar inside and outside the clause, 17-60. Cambridge: Cambridge University Press.

������ �����2�"�������� ��

The probabilistic behaviour of GO and COME as conjoining verbs in English

corresponds to

The categorical behaviour of GO and COME in the serial verb construction of some other languages

���� �

� What kinds of verbs occur in the V and V construction?

� Are there special meanings/ grammaticalizations which arise in the V andV construction?

� How are V1 and V2 integrated semantically?

����� � ��� ����"��!��

• then you try and comprehend it in maori eh

• for years basically he had to try and convince the politicians

• i might want to try and catch up with him again

Distribution of try and V in Wellington Corpus

0

0

1*

36

130

Spoken

0tried and VVD

0tries and VVZ

0trying and VVG

2try and VVB

16try and VVI

Written

* Ted kept trying and scraping all this paint off the inside

3������(������ �

� The distribution of the combination is skewed across inflectional categories, i.e., it has a defective paradigm in this construction

� More frequent in spoken than in written corpus

� The meaning of the whole is unified� try and V is grammaticalized

���� �

� What kinds of verbs occur in the V and V construction?

� Are there special meanings/ grammaticalizations which arise in the V andV construction?

� How are V1 and V2 integrated semantically?

��������!����(�-+��� �

GO and TELL

GO and VISIT

GO and PROVE ME WRONG (Colloquial)

��������!����(�-+�����

GO and TELL

GO and VISIT

GO and PROVE ME WRONG (Colloquial)

-+

LM

TR

TIME

Detailed view*

TR

TIME

Schematic view

* Langacker, R. W. (1988). A view of linguistic semantics. In B. Rudzka-Ostyn (Ed.), Topics in Cognitive Linguistics, 49-90. Amsterdam and Philadelphia: John Benjamins. p. 62

-+��� �����/��������0

TIME

GO TELL

TR

GO and TELL in Spatio-temporal Domain

-+��� �����/��������0

TIME

TELL

TR

GO and TELL in Intentional Domain

"��! �4�����4��(�-+��� �����/��������0

TIME

GO TELL

TR

GO and TELL in Spatio-temporal and Intentional Domains

"(2����! �4�����4��(������

SPACE CUTTING SILVERWARE

COMPLEX MATRIX

KNIFE

Langacker, R. W. (1988). A view of linguistic semantics. In B. Rudzka-Ostyn (Ed.), Topics in Cognitive Linguistics, 49-90. Amsterdam and Philadelphia: John Benjamins. p. 57

-+�������!��������*�

� Fa d’Ambu (Portuguese-based creole, Equatorial Guinea)ba ‘go’ > ‘(in order) to’

� Shona (Bantu, Zimbabwe)(ku-)enda ‘to go’ > -ndo- purpose marker

� Rama (Chibchan, Nicaragua, virtually extinct)bang ‘go’ > -bang subordinating conjunction of goal, purpose

Heine, B. and T. Kuteva (1999, ms.) Common grammaticalization processes in the languages of the world.

��������!����(�-+�����

GO and TELL

GO and VISIT

GO and PROVE ME WRONG (Colloquial)

���� �

��!������

5���( �6 ���5$���6

Langacker, R. W. (1988). A view of linguistic semantics. In B. Rudzka-Ostyn (Ed.), Topics in Cognitive Linguistics, 49-90. Amsterdam and Philadelphia: John Benjamins. p. 59

Profile = hypotenuse

Base = whole triangle

�����

� visit. v.t. 1. to go to see (a person, place, etc.) in the way of friendship, ceremony, duty, business, curiosity or the like. [The Macquarie Dictionary]

� visit. 1. v.t. go or come to see (person, place, etc. or abs.) as act of friendship or ceremony, on business or for a purpose [Concise Oxford Dictionary]

�����

� visit. v.t. 1. to go to see (a person, place, etc.) in the way of friendship, ceremony, duty, business, curiosity or the like. [The Macquarie Dictionary]

� visit. 1. v.t. go or come to see (person, place, etc. or abstract) as act of friendship or ceremony, on business or for a purpose [Concise Oxford Dictionary]

�������7"

1. yes but i go all over cambridge visit the old haunts go out to our

2. cos you know she goes up to auckland to visit her father a all the time and he can give her p

3. yeah all i can do i'm going going in to visit rang see rangi yeah oh that's

4. anything you know there's no popping around to visit somebody in the evening

5. we come back to our waka and then paddle downto moehau and visit all the other marae in waima

-+��� �&7&��/��������0

TIME

GO VISIT

TR

GO and VISIT

in Spatio-temporal Domain

"��! �4�����4��(�-+��� �&7&��/��������0

TIME

GO VISIT

TR

GO and VISIT in Spatio-temporal and Intentional Domains

���� ������! �������8

� they go and pick her up at ten thirty� they go and see a solicitor� The team prays for two hours before they go

and return later to pray again.

��������!����(�-+�����

GO and TELL

GO and VISIT

GO and PROVE ME WRONG (Colloquial)

"� ���� ���� �

1. so i i went and told this guy 2. the silly council went and sprayed the side of the

banks3. so she went and moved us up to new plymouth4. and then a day later he went and wrote this poem

called thine5. go on go and prove me wrong now 6. the punters they went and paid forty odd dollars

to win

"� ���� �-+������+���,����+)-

/��������0

TIME

ACT (non-specific

PROVE ME WRONG

TR

GO and PROVE ME WRONG in Spatio-temporal Domain

"��! �4�����4��(��� ���� �-+����

��+���,����+)- /��������0

TIME

ACT (non-specific

PROVE ME WRONG

TR

GO and PROVE ME WRONG in Spatio-temporal and Intentional Domains

-+��� �

� Motion is present in varying degrees� Always purposeful human motion in the

Wellington Corpus� Exhibits a range in types of semantic

integration

7������

V and V is asymmetrical in the types of verbs that occur as V1 and V2 (quite apart from any “reversal of predicates” test a laHuddleston and Pullum)

MOTION verbs account for most instances of V1and in infinitival, base, and past tense forms

7�������/���20

The most common event schema encoded by V and V is

SUBSEQUENT ACTIVITYAT END OF PATH

+MOTION ALONG A PATH

7�������/���20

V and V is worthy of study:

� frequent, therefore a large amount of data can be observed

� a range of integration types (even when the focus is just on GO and V)

� completed and emergent grammaticalizations are evident