-
Notifications
You must be signed in to change notification settings - Fork 0
/
formgram-lfg.tex
1031 lines (755 loc) · 23.7 KB
/
formgram-lfg.tex
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
\section{Lexical Functional Grammar (LFG)}
\subtitle{Lexical Functional Grammar (LFG)}
\huberlintitlepage[22pt]
\outline{
\begin{itemize}
\item Introduction and basic terms
\item Phrase structure grammar and \xbar Theory
\item Government \& Binding (GB)
\item {Generalized Phrase Structure Grammar (GPSG)}
\item {Feature descriptions, feature structures and models}
\item \alert{Lexical Functional Grammar (LFG)}
%\item PATR
\item Categorial Grammar (CG)
\item Head-Driven Phrase Structure Grammar (HPSG)
%\item Konstruktionsgrammatik (CxG)
\item Tree Adjoning Grammar (TAG)
\end{itemize}
%\tableofcontents
}
\frame{
\frametitle{Reading material}
\citew[Chapter~7]{MuellerGT-Eng} (without 7.1.5 on semantics)
}
\exewidth{(123)}
\frame{
\frametitle{Lexical Functional Grammar (LFG)}
\begin{itemize}[<+->]
\item Developed by Joan Bresnan and Ron Kaplan in the 1980s.\nocite{BK82a}
\item LFG is part of so-called West-Coast-Linguistics:\\
Joan Bresnan (LFG) and Ivan Sag (HPSG) did their PhD with Chomsky\\
(MIT is situated at the East Coast of the US,\\
while Stanford, Palo Alto and Berkeley are in the Bay Area in California)
\item LFG aims for psycholinguistical plausibility and wants to be implementable
\item teaching material and overview articles:
\citew{BATW2016a,Dalrymple2006a}
\item In-depth works on German: \citew{Berman96a-u,Berman2003a} and \citew{Cook2001a}
\end{itemize}
}
\subsection{General remarks on the representational format}
\frame{
\frametitle{General remarks on the representational format}
\begin{itemize}
\item multiple levels of representation:
\begin{itemize}
\item c-structure (constituent structures, licensed by PSG, \xbar structures)
\pause
\item f-structure (functional structure)
\end{itemize}
\pause
\item Mappings relate c- and f-structure.
\end{itemize}
}
\subsubsection{Functional structure}
\subsubsubsection{Grammatical functions}
\frame{
\frametitle{Grammatical functions and f-structure}
\begin{itemize}
\item In LFG, grammatical functions (subject, object, \ldots) play a very important role.\\
They are primitives of the theory.
\pause
\item A sentence such as (\mex{1}a) has the functional structure in (\mex{1}b):
\eal
\ex David devoured a sandwich.
\ex \lfgms{ \gruen<3>{pred} & `DEVOUR\sliste{\gruen<4>{\lfgsubj},\gruen<4>{\lfgobj}}'\\
subj & \lfgms{ \gruen<3>{pred} & `DAVID' \\
}\\
obj & \lfgms{ spec & A\\
\gruen<3>{pred} & `SANDWICH'\\
}\\
}
\zl
\pause
\item All lexical items that have a meaning (\eg nouns, verbs, adjectives) contribute a
\textsc{pred} feature with a corresponding value.
\pause
\item
The grammatical functions governed by a head (government = subcategorization)
are determined in the specification of \textsc{pred}.
\end{itemize}
}
\frame{
\frametitle{Governable grammatical functions}
The respective grammatical functions are called \emph{governable grammatical functions}.
Examples:
\begin{tabular}[t]{@{}lp{30em}@{}}
\sc subj: & subject \\
\pause
\sc obj: & object\\
\pause
%% \sc comp: & Satz- oder abgeschlossenes (nicht-prädikatives) Infinitivkomplement\\
\sc comp & sentential complement\\
%% \sc xcomp: & offenes (prädikatives) Komplement, oft infinitivisch, dessen {\sc subj}"=Funktion
%% extern kontrolliert wird\\
\objtheta: & secondary \textsc{obj} functions that are related to a special, language \\
& specific set of grammatical roles; English has \objtheme only.\\
\pause
%
\obltheta: & a group of thematically restricted oblique functions, as for instance
{\obl\downlett{GOAL}} or {\obl\downlett{AGENT}}. These often correspond to adpositional
phrases in c-structure.\\
\end{tabular}
}
\frame{
\frametitle{Non-governable grammatical functions}
Apart from this there are non-governable grammatical functions.
Examples:
\begin{tabular}[t]{@{}lp{26em}@{}}
\textsc{adj}: & adjuncts \\
%
\textsc{topic}: & the topic of an utterance\\
%
\textsc{focus}: & the focus of an utterance\\
\end{tabular}
}
\subsubsubsection{Functional descriptions}
\frame{
\frametitle{Functional descriptions}
Reference to a value of the feature \textsc{tense} in the functional structure $f$:
\ea
($f$ \lfgtense)
\z
\pause
It is possible to say something about the value which this feature should have in the feature description.
\ea
($f$ \lfgtense) = \lfgpast
\z
\pause
The value of a feature may also be a specific f"=structure.\\
(\mex{1}) ensures that the \subjf in $f$ is the f"=structure $g$:
\ea
($f$ \lfgsubj) = $g$
\z
}
\frame[shrink]{
\frametitle{Descriptions and f-structures}
\eal
\ex David sneezed.
\ex
\begin{tabular}[t]{l}
($f$ \lfgpred) = {\small `SNEEZE\sliste{\lfgsubj}'}\\
($f$ \lfgtense) = {\small PAST}\\
($f$ \lfgsubj) = $g$\\
($g$ \lfgpred) = {\small `DAVID'}
\end{tabular}
\zl
\pause
The description in (\mex{0}b) describes the following structure:
\ea
$f$: \lfgms{ pred & `SNEEZE\sliste{\lfgsubj}'\\
tense & \lfgpast\\
subj & $g$: \onems{ pred `DAVID' }\\
}
\z
(\mex{-1}b) also describes many other structures which contain further features.\\
We are only interested in minimal structures containing the information provided in the description.
}
\exewidth{(100)}
\frame{
\frametitle{Mappings from c-structure to f-structure}
\eal
\ex David sneezed.
\ex
\tree{IP}{%
\tree[b]{NP}{\tree{N$'$}{\tree{N}{\le{David}}}}
\tree{I$'$}{\tree{VP}{\tree{V$'$}{\tree{V}{\le{sneezed}}}}}}%
\hspace*{4em}%
\raisebox{-2em}{\lfgms{
pred & `SNEEZE\sliste{\lfgsubj}'\\
tense & PAST\\
subj & \rnode{i}{\lfgms{ pred & `DAVID' \\
}}\\
}}\\
\ltor[-15]{b}[175]{i}
\Aput*{$\phi$}
\zl
A phrase and its head always correspond to the same f"=structure.\\
IP, I$'$ and I (and also VP) are mapped onto the same f"=structure.
}
\frame{
\frametitle{Heads and f-structure}
A phrase and its head always correspond to the same f"=structure:
\ea
\begin{tabular}[t]{@{}c@{}}
\rnode{a}{\rnode{v1}{V$'$}}\\[2ex]
\rnode{b}{\rnode{v}{V}}\\[2ex]
\rnode{sneezed}{sneezed}\\
\end{tabular}
\hspace*{4em}
\rnode{d}{\raisebox{-2em}{\fd{\fdand{\feat{\lfgpred}{\small `SNEEZE\sliste{\lfgsubj}'}
\feat{\lfgtense}{\small PAST}}}}}
\ncline{v1}{v}\ncline{v}{sneezed}%
\ltor{a}{d}
\Aput*{$\phi$}
\ltor{b}{d}
\z
}
\frame{
\frametitle{IP, I$'$, I and VP are mapped to the same f-structure}
In LFG grammars of English, the CP/IP system is assumed as in \gbt.\\
IP, I$'$ and I (and also VP) are mapped onto the same f"=structure.
\eal
\ex David is yawning.
\ex {\tree[a]{IP}{%
\tree{NP}{\tree{N$'$}{%
\tree{N}{\le{\em David}}}}
\tree[b]{I$'$}{%
\tree[c]{I}{\le{\em is}}
\tree[d]{VP}{\tree[e]{V$'$}{\tree[f]{V}{\le{\em yawning}}}}}}}%
\hspace*{4em}%
{\rnode{o}{\raisebox{-2em}{\fd{\fdand{\feat{\lfgpred}{\small `YAWN\sliste{\lfgsubj}'}
\feat{\lfgtense}{\small PRES}
\feat{\lfgsubj}{\fdand{\feat{\lfgpred}{\small `DAVID'}}}}}}}}
\ltor{a}{o}
\ltor{b}{o}
\ltor[10]{c}{o}
\ltor{d}{o}
\ltor{e}{o}
\ltor{f}{o}
\zl
}
%% \subsubsection{Funktionale Eindeutigkeit ({\em Functional Uniqueness})}
%% \frame{
%% \frametitle{Funktionale Eindeutigkeit ({\em Functional Uniqueness})}
%% }
\subsubsection{Completeness}
\frame{
\frametitle{Completeness}
Elements required in the \textsc{pred} value have to be realized.
\eal
\ex[*]{David devoured.
}
\ex[]{
\lfgms{ pred & \small `DEVOUR\sliste{\gruen{\lfgsubj},\rot{\lfgobj}}'\\
\gruen{subj} & \lfgms{ pred & `DAVID' \\
}\\
}
}
\zl
\textsc{obj} is missing a value in (\mex{0}b), which is
why (\mex{0}a) is ruled out by the theory.
}
\subsubsection{Coherence}
\frame[shrink]{
\frametitle{Coherence}
All argument functions in a given f"=structure have to be selected in the value of the local
\textsc{pred} attribut.
\eal
\ex[*]{
David devoured a sandwich that Peter sleeps.%\\
%`David verschlang ein Sandwich, dass Peter schläft.'
}
\ex[]{
\lfgms{ pred & \small `DEVOUR\sliste{\gruen{\lfgsubj},\gruen{\lfgobj}}'\\
\gruen{subj} & [ \textsc{pred} `DAVID' ] \\
\gruen{obj} & \lfgms{ spec & A\\
pred & `SANDWICH'\\
}\\
\rot{comp} & \lfgms{ pred & `SLEEP\sliste{\lfgsubj}'\\
subj & \lfgms{ pred & `PETER'\\
}\\
}
}
}
\zl
(\mex{0}a) is ruled out because \textsc{comp} does not appear under the arguments of \emph{devour}.
}
\subsubsection{Restrictions on the c-structure/f-structure relation}
\frame{
\frametitle{Restrictions on the c-structure/f-structure relation}
$\uparrow$ : the f"=structure of the immediately dominating node\\
$\downarrow$ : f-structure of the c"=structure node bearing the annotation
\ea
V$'$ $\to$ \begin{tabular}[t]{@{}r@{~=~}l@{}}
\multicolumn{2}{@{}l@{}}{\hspaceThis{f-structure of the mother~}V}\\
$\uparrow$ & $\downarrow$\\
f-structure of the mother & own f-structure\\
\end{tabular}
\z
\ea
\talltree[a]{V$'$}{\le[b]{V}}%
\hspace*{3em}%
\rnode{d}{[\ ]}
\ltor{a}{d}
\ltor{b}{d}
\z
}
\frame{
\frametitle{V$'$ rule with object}
\ea
\phraserule{V$'$}{
\rulenode{V\\* \up~=~\down}
\rulenode{NP\\*(\up\ \lfgobj) = \down}}
\z
\ea
\talltree[a]{V$'$}{\le[b]{V} \le[c]{NP}}%
\hspace*{3em}%
\rnode{d}{\fd{\fdand{\feat{\lfgobj}{\rnode{e}{[\ ]}}}}}
\ltor{a}{d}
\ltor[20]{b}{d}
\ltor{c}[190]{e}
\z
\bigskip
annotation on the NP:\\
the \textsc{obj} value in the f"=structure of the mother
\mbox{(\up\ \lfgobj)} is identical\\to the f"=structure of the NP node (\down).
}
\frame{
\frametitle{A lexical entry}
Similarly in lexical entries:
\ea
\catlexentry{sneezed}{V}{(\up\ \lfgpred) = {\small `SNEEZE\sliste{\lfgsubj}'}\\*
(\up\ \lfgtense) = \small PAST}
\z
\ea
\tree{V}{\le{sneezed}}
\hspace*{4em}
\rnode{d}{\mbox{\lfgms{ pred & {\small `SNEEZE\sliste{\lfgsubj}'}\\
tense & \lfgpast}}}
\ltor{top}{d}
\z
}
\subsection{Passive}
\subsubsection{Lexical Integrity}
\frame{
\frametitle{Lexical Integrity}
\begin{itemize}
\item \citet{BM95a}:\\
Words are atoms of syntactic structure.\\
Syntactic rules cannot create new words or make reference to the internal structure of words.
\pause
\item Every terminal node (each ``leaf'' of the tree) is a word.
\pause
\item
This means: Pollock's \citeyearpar{Pollock89a-u} analysis of (\mex{1}) is excluded:
\ea
\gll Marie ne parl-er-ait pas \\
Marie \textsc{neg} speak-\textsc{cond}-\textsc{3sg} \textsc{neg}\\
\glt `Marie would not speak.'
\z
In Pollock's analysis, the various morphemes are in specific positions in the tree and are combined only after certain movements have been carried out.
\end{itemize}
}
\frame{
\frametitlefit{GB analysis with morphemes as terminal symbols (Pollock 1989)}
\vfill
\centerline{%
\scalebox{0.5}{
\begin{forest}
for tree={parent anchor=south, child anchor=north,align=center,base=bottom}
[AgrP
[Spec-AgrP,name=specagr]
[Agr$'$
[Agr
[\textit{-ait},name=ait]]
[NegP
[Spec-NegP
[\textit{pas},name=pas]]
[Neg$'$
[Neg
[\textit{ne},name=ne]]
[TP
[Spec-TP]
[T$'$
[T
[\textit{-er-},name=er]]
[VP
[Spec-VP
[\textit{Marie},name=marie]]
[V$'$
[V
[\textit{parl-},name=parl]]]]]]]]]]
\begin{pgfinterruptboundingbox}% otherwise the picture gets larger due to the control points
\draw[->,dotted] (parl.south west) .. controls +(225:1cm) and +(south:0.4cm) .. (er.south);
\draw[->,dotted] (er.south west) .. controls +(left:1cm) and +(south:0.4cm) .. (ne.south);
\draw[->,dotted] (ne.south west) .. controls +(left:1cm) and +(south:0.4cm) .. (ait.south);
\draw[->,dotted] (marie.-90) .. controls +(225:6cm) and +(250:3cm) .. (specagr.-90);
\end{pgfinterruptboundingbox}
\end{forest}
}%scalebox
}
\vfill
\hfill
\gll Marie ne parl-er-ait pas \\
Marie \textsc{neg} speak-\textsc{cond}-\textsc{3sg} \textsc{neg}\\\hfill\mbox{}
\vfill
}%frame
\subsubsection{Lexical integrity and passive}
\frame[shrink=5]{
\frametitle{Lexical integrity and passive (I)}
\begin{itemize}
\item observation: there are passivized adjectives which show the same morphological idiosyncrasies as the corresponding participles \citep[\page 31]{Bresnan2001a}
\eal
\ex a well-written novel (write -- written)
\ex a recently given talk (give -- given)
\ex my broken heart (break -- broken)
\ex an uninhabited island (inhabit -- inhabited)
\ex split wood (split -- split)
\zl
\pause
\item The adjectival participles have passive argument structure: the subject is suppressed and the
object is what is predicated over (the noun):
\eal
\ex Aicke broke my heart.
\ex My heart is broken.
\ex my \alert{broken heart}
\zl
\eal
\ex My friend is smart.
\ex my \alert{smart friend}
\zl
\end{itemize}
}
\frame{
\frametitle{Lexical integrity and passive (II)}
\begin{itemize}
\item Passive participle and adjectival participle have the same form:
\eal
\ex Aicke broke my heart.
\ex My heart was \alert{broken}.
\ex my \alert{broken} heart
\zl
\pause
\item If one assumes lexical integrity,\\
then adjectives have to be derived in the lexicon.
\pause
\item If the verbal passive were not a lexical process, but rather a phrase"=structural one, then
the form identity would remain unexplained.
\end{itemize}
}
\subsubsection{Passive as a lexical process}
\frame{
\frametitle{Passive as a lexical process}
\begin{itemize}[<+->]
\item Grammatical functions are primitives of the theory.\\
(that is not derived from tree positions [\eg subject = SpecIP])
\item Words (that is, fully inflected word forms) determine grammatical functions of their arguments.
\item There is a hierarchy of grammatical functions.
\item When participles are formed in morphology, the highest argument is suppressed.
\item The next-highest argument is not realized as OBJECT but as SUBJECT.
\end{itemize}
}
\frame{
\frametitle{The lexical rule}
\begin{itemize}
\item The assignment of grammatical functions is regulated by the \alert{Lexical
Mapping Theory}.
\pause
\item Earlier works \citep{Bresnan82a} had an explicit formulation of the passive rule:
\ea
Passive rule:\\
\begin{tabular}{@{}l@{~$\mapsto$~}l@{}}
\gruen<2>{(SUBJ)} & \gruen<2>{$\varnothing$/(OBL)}\\
\gruen<3>{(OBJ)} & \gruen<3>{(SUBJ)}
\end{tabular}
\z
This means: The subject is either not expressed at all ($\varnothing$) or\\
as oblique Eelement (as a \emph{von}"=PP in German)
\pause
If there is an accusative object, this will be realized as subject.
\end{itemize}
}
\subsection{Verb position}
\frame[shrink=5]{
\frametitle{Verb position}
\begin{itemize}
\item two options:
\begin{itemize}
\item a trace in verb"=final position (as in GB\indexgb) (see \citealp{Choi99a-u},
\citealp[Section~2.1.4]{Berman96a-u}) and
\pause
\item so"=called \alert{extended head domains} (see \citealp{Berman2003a}).
\end{itemize}
\pause
\item Extended head domains: verb is simply omitted in the verb phrase:
\ea
VP $\to$ NP* (V)\hspace{2cm} (preliminary version)
\z
All parts of the VP are optional (indicted by brackets and Kleene star).
\pause
\item As in GB analyses, the verb is in the C position.\\
It contributes f-structure informtion from there.
\pause
\item VP without V????\\
We have to make sure that all necessary items are present and nothing more:\\
coherence and completeness.
Where the necessary information for this comes from is not important.
\end{itemize}
}
\frame{
\frametitle{An example of the verb placement analysis}
\bigskip
\centerline{\includegraphics[width=.75\textwidth]{Figures/verschlingt-david-den-apfel-lfg-lsp-crop}}
\bigskip
\centerline{Analysis adapted from \citet[\page 41]{Berman2003a}.}
\vfill
}
\subsection{Local reordering}
\frame{
\frametitle{Local reordering}
\begin{itemize}
\item Two options are discussed:
\begin{itemize}
\item movement of arguments from a base configuration as in GB (see \citealp{Choi99a-u})
\pause
\item direct licensing by phrase structure rules (see Berman \citeyear[Section~2.1.3.1]{Berman96a-u}; \citeyear{Berman2003a})
\end{itemize}
\end{itemize}
}
\frame{
\frametitle{Local reordering as ``base generateion'' (I)}
\begin{itemize}
\item Case requirements are specified in lexical items:
\ea
\label{le-verschlingen}
\catlexentry{verschlingt}{V}{(\up\ \lfgpred) = {\small `VERSCHLINGEN\sliste{\lfgsubj,\lfgobj}'}\\*
(\up\ \lfgsubj{} {\small AGR CAS}) = NOM\\*
(\up\ \lfgobj{} {\small AGR CAS}) = ACC\\*
(\up\ \lfgtense) = \small PRES}
\z
\pause
\item GPSG: all arguments are combined with the head in one go.
\pause
\item LFG: no argument is combined with the verb and we get a VP without anything.
\ea
\label{LFG-v-vp}
\phraserule{VP}{
\rulenode{(V)\\* \up~=~\down}}
\z
\pause
\item Hm.
\pause
\item But this is just to get the recursion going.
\end{itemize}
}
\frame{
\frametitle{Local reordering as ``base generateion'' (II)}
\begin{itemize}
\item Case requirements are specified in lexical items:
\ea
\label{le-verschlingen}
\catlexentry{verschlingt}{V}{(\up\ \lfgpred) = {\small `VERSCHLINGEN\sliste{\lfgsubj,\lfgobj}'}\\*
(\up\ \lfgsubj{} {\small AGR CAS}) = NOM\\*
(\up\ \lfgobj{} {\small AGR CAS}) = ACC\\*
(\up\ \lfgtense) = \small PRES}
\z
\ea
\label{LFG-v-vp}
\phraserule{VP}{
\rulenode{(V)\\* \up~=~\down}}
\z
\pause
\item Recursive rule to add NP arguments:
\ea
\label{lfg-vp-regel}
\phraserule{VP}{
\rulenode{NP\\* (\upsp \lfgsubj|\lfgobj|\objtheta) = \down}
\rulenode{VP\\* \up~=~\down}}
\z
\pause
\item similar rules for PP arguments and so on.
\end{itemize}
}
\frame{
\frametitle{Binary branching with normal order (nom, acc)}
\bigskip
\centerline{%
\includegraphics{Figures/david-den-apfel-verschlingt-lfg-lsp-crop}
}
\bigskip
%\centerline{Analysis adapted from \citet[\page 41]{Berman2003a}.}
\vfill
}
\frame{
\frametitle{Binary branching with marked order (acc, nom)}
\bigskip
\centerline{%
\includegraphics{Figures/den-apfel-david-verschlingt-lfg-lsp-crop}
}
\bigskip
%\centerline{Analysis adapted from \citet[\page 41]{Berman2003a}.}
\vfill
}
\subsection{Long"=distance dependencies}
\subsubsection{Discourse functions}
\frame{
\frametitle{Long"=distance dependencies: Discourse functions (I)}
\begin{itemize}
\item Observation: the displaced constituent \emph{Chris} is characterized by two functions:
\ea
Chris, we think that David saw.
\z
\begin{itemize}
\item an \alert{argument function} which is normally realized in a different position:\\
the \lfgobj function of \emph{saw}
\pause
\item a certain emphasis of the information"=structural status in this construction:\\
\textsc{topic} in the matrix clause -- a \alert{discourse function}
\end{itemize}
\end{itemize}
}
\frame{
\frametitle{Discourse functions (II)}
\begin{itemize}
\item grammaticalized discourse functions: \textsc{topic} and \textsc{focus}\\
(\textsc{subj} is a default discourse function).
\begin{itemize}
\item Only \alert{grammaticalized} discourse functions are represented on the level of f"=structure, that is, those
that are created by a fixed syntactic mechanism and that interact with the rest of the syntax.
%% (F\"ur die umfassende Repr\"asentation von informationsstrukturellen
%% Eigenschaften wird in LFG eine separate Informationsstruktur
%% angenommen.)
\pause
\item \textsc{topic} and \textsc{focus} are not lexically subcategorized and are therefore not
subject to the completeness and coherence conditions.
\pause
\item \textsc{topic} and \textsc{focus} are identified with an
f"=structure that bears an argument function.
\end{itemize}
\end{itemize}
}
\frame{
\frametitle{Discourse functions in f-structure}
\smallframe
\eal
\ex Chris, we think that David saw.
\ex
\lfgms{ pred & `THINK\sliste{\lfgsubj,\comp}' \\
topic & \rnode{topic}{\lfgms{ pred & `CHRIS' \\
}}\\[4mm]
subj & \lfgms{ pred & `pro'\\
}\\
comp & \lfgms{ pred & `SEE\sliste{\lfgsubj,\lfgobj}\\
subj & \lfgms{ pred & `DAVID' \\
}\\
obj & \rnode{obj}{}\\
}\\
}
\nccurve[ncurv=2.2]{topic}{obj}
\zl
\bigskip
The line means: The value of \textsc{topic} is identical to \textsc{comp obj}.
The constraint: (\up \textsc{topic})=(\up \textsc{comp obj})
}
\frame{
\frametitle{Different levels of embedding (I)}
\eal
\ex Chris, we saw.
\ex
\lfgms{ pred & `SEE\sliste{\lfgsubj,\lfgobj}' \\
topic & \rnode{topic}{\lfgms{ pred & `CHRIS' \\
}}\\[4mm]
subj & \lfgms{ pred & `pro'\\
}\\
obj & \rnode{obj}{}\\
}
%\nodecurve[r]{topic}[r]{obj}{15em}
\nccurve[nodesepA=1pt,ncurv=2.2]{topic}{obj}
\zl
\bigskip
The constraint: (\up \textsc{topic})=(\up \textsc{obj})
}
\frame{
\frametitle{Different levels of embedding (II)}
\smallframe
\eal
\ex Chris, we think Anna claims that David saw.
\ex
\scalebox{0.85}{%
\lfgms{ pred & `THINK\sliste{\lfgsubj,\comp}' \\
topic & \rnode{topic}{\lfgms{ pred & `CHRIS' \\
}}\\[4mm]
subj & \lfgms{ pred & `pro'\\
}\\
comp & \lfgms{ pred & `CLAIM\sliste{\lfgsubj,\comp}\\
subj & \lfgms{ pred & `ANNA' \\
}\\
comp & \lfgms{ pred & `SEE\sliste{\lfgsubj,\lfgobj}\\
subj & \lfgms{ pred & `DAVID' \\
}\\
obj & \rnode{obj}{}\\
}\\
}\\
}
%\nodecurve[r]{topic}[r]{obj}{15em}%
\nccurve[ncurv=2.2]{topic}{obj}
}
\zl
The constraint: (\up \textsc{topic})=(\up \textsc{comp comp obj})
}
\subsubsection{Functional uncertainty}
\frame{
\frametitle{Functional uncertainty}
\begin{itemize}
\item The constraints are c-structure constraints:
\ea
\begin{tabular}[t]{@{}ccc@{~=~}lc@{}}
CP & $\rightarrow$ & \multicolumn{2}{l}{\hspaceThis{(\up \textsc{topic})}XP} & C$'$ \\
& & (\up \textsc{topic}) & \down & \up=\down \\
& & (\up \textsc{topic}) & (\up \textsc{comp obj})\\
\end{tabular}
\z
\pause
\item But we have different levels of embedding:
\ea
(\up \textsc{topic})=(\up \textsc{obj})\\
(\up \textsc{topic})=(\up \textsc{comp obj})\\
(\up \textsc{topic})=(\up \textsc{comp comp obj})\\
\ldots
\z
\pause
\item The generalization over these equations is:
\ea
(\up \textsc{topic})=(\up \textsc{comp* obj})
\medskip
\z
The Kleene star `*' stands for arbitrarily many repetitions of \comp.
\end{itemize}
}
\frame{
\frametitle{Disjunctions and variables for grammatical functions}
\begin{itemize}[<+->]
\item The fronted element is not necessarily a \textsc{topic},
\focus is possible as well.
\item It is possible to state disjunctions:
\ea
(\up \textsc{topic$|$focus})=(\up \textsc{comp* obj})
\z
\item \textsc{topic$|$focus} can be abbreviated by using the shortcut \textsc{df} (discourse function).
%% \begin{tabular}{@{}ccc@{~=~}lc@{}}
%% CP & $\rightarrow$ & \multicolumn{2}{l}{\hspaceThis{(\up \textsc{topic})}XP} & C$'$ \\
%% & & (\up \textsc{DF}) & \down & \up=\down \\
%% & & (\up \textsc{DF}) & (\up \textsc{comp* GF})\\
%% \end{tabular}
\end{itemize}