Paraphrases

ERG
#1

Why does the grammar paraphrase a prop sentence ([ e SF: prop ]) to a question?

% ace -g ~/hpsg/sick/erg.dat -Tf1 | ace -g ~/hpsg/sick/erg.dat -e
Before the scientists hid the artist waited
NOTE: 1 readings, added 2164 / 478 edges to chart (182 fully instantiated, 144 actives used, 125 passives used)	RAM: 6315k
Before the scientists hid, the artist waited.
The artist waited before the scientists hid?
The artist waited before the scientists hid.
The artist waited before the scientists hid.
Before the scientists hid the artist waited.
The artist waited, before the scientists hid.
NOTE: 692 passive, 760 active edges in final generation chart; built 841 passives total. [6 results]
#2

Are you sure that your input sentence is [ SF prop ] and not [ SF prop-or-ques ]? What happens if you give it final punctuation (i.e. a period)?

#3

The command above does the paraphrase from the first reading, right? If I asked only for the analysis:

% ace -g ~/hpsg/sick/erg.dat -Tf1
Before the scientists hid the artist waited
SENT: Before the scientists hid the artist waited
[ LTOP: h0
INDEX: e2 [ e SF: prop ]
RELS: < [ _before_x_h<0:6> LBL: h1 ARG0: e2 ARG1: h4 ARG2: h5 ]
 [ _the_q<7:10> LBL: h6 ARG0: x7 [ x PERS: 3 NUM: pl IND: + ] RSTR: h8 BODY: h9 ]
 [ _scientist_n_1<11:21> LBL: h10 ARG0: x7 ]
 [ _hide_v_1<22:25> LBL: h11 ARG0: e12 [ e SF: prop TENSE: past MOOD: indicative ] ARG1: x7 ARG2: i13 ]
 [ _the_q<26:29> LBL: h14 ARG0: x3 [ x PERS: 3 NUM: sg IND: + ] RSTR: h15 BODY: h16 ]
 [ _artist_n_1<30:36> LBL: h17 ARG0: x3 ]
 [ _wait_v_1<37:43> LBL: h18 ARG0: e19 [ e SF: prop TENSE: past MOOD: indicative PROG: - PERF: - ] ARG1: x3 ARG2: i20 ] >
HCONS: < h0 qeq h1 h4 qeq h18 h5 qeq h11 h8 qeq h10 h15 qeq h17 >
ICONS: < > ]
NOTE: 1 readings, added 2164 / 478 edges to chart (182 fully instantiated, 144 actives used, 125 passives used)	RAM: 6315k

Before the scientists hid the artist waited.
SENT: Before the scientists hid the artist waited.
[ LTOP: h0
INDEX: e2 [ e SF: prop ]
RELS: < [ _before_x_h<0:6> LBL: h1 ARG0: e2 ARG1: h4 ARG2: h5 ]
 [ _the_q<7:10> LBL: h6 ARG0: x7 [ x PERS: 3 NUM: pl IND: + ] RSTR: h8 BODY: h9 ]
 [ _scientist_n_1<11:21> LBL: h10 ARG0: x7 ]
 [ _hide_v_1<22:25> LBL: h11 ARG0: e12 [ e SF: prop TENSE: past MOOD: indicative ] ARG1: x7 ARG2: i13 ]
 [ _the_q<26:29> LBL: h14 ARG0: x3 [ x PERS: 3 NUM: sg IND: + ] RSTR: h15 BODY: h16 ]
 [ _artist_n_1<30:36> LBL: h17 ARG0: x3 ]
 [ _wait_v_1<37:43> LBL: h18 ARG0: e19 [ e SF: prop TENSE: past MOOD: indicative PROG: - PERF: - ] ARG1: x3 ARG2: i20 ] >
HCONS: < h0 qeq h1 h4 qeq h18 h5 qeq h11 h8 qeq h10 h15 qeq h17 >
ICONS: < > ]
NOTE: 1 readings, added 2461 / 509 edges to chart (209 fully instantiated, 143 actives used, 153 passives used)	RAM: 6497k

That is, both are [ e SF: prop ] with and without the final period.