WEBVTT 00:00:00.000 --> 00:00:01.952 align:middle line:90% [SQUEAKER] 00:00:01.952 --> 00:00:02.928 align:middle line:90% [SQUEAK] 00:00:02.928 --> 00:00:04.392 align:middle line:90% [SQUEAKIN] 00:00:04.392 --> 00:00:11.192 align:middle line:90% 00:00:11.192 --> 00:00:12.900 align:middle line:84% NORVIN RICHARDS: I'm going to start today 00:00:12.900 --> 00:00:14.880 align:middle line:84% with a little bit of a review of something 00:00:14.880 --> 00:00:18.690 align:middle line:84% that we talked about sort of earlier 00:00:18.690 --> 00:00:21.540 align:middle line:84% when we were doing syntax because it came in handy. 00:00:21.540 --> 00:00:23.520 align:middle line:84% At a certain point, you guys were pushing me 00:00:23.520 --> 00:00:26.190 align:middle line:84% with questions about the fine details 00:00:26.190 --> 00:00:27.300 align:middle line:90% of syntactic structures. 00:00:27.300 --> 00:00:29.520 align:middle line:84% And I wasn't able to answer your questions 00:00:29.520 --> 00:00:32.400 align:middle line:84% without making reference to another kind of test 00:00:32.400 --> 00:00:35.260 align:middle line:84% for syntactic structure, which was binding theory. 00:00:35.260 --> 00:00:37.200 align:middle line:84% And so I showed you binding theory back then, 00:00:37.200 --> 00:00:40.160 align:middle line:84% but I'm going to go over it again in a little more detail. 00:00:40.160 --> 00:00:42.660 align:middle line:84% And we'll talk about some things that we didn't get a chance 00:00:42.660 --> 00:00:44.530 align:middle line:90% to talk about last time. 00:00:44.530 --> 00:00:46.470 align:middle line:90% And then we'll go on from there. 00:00:46.470 --> 00:00:48.750 align:middle line:84% Binding theory is the theory that's 00:00:48.750 --> 00:00:52.170 align:middle line:84% meant to handle facts like the facts that 00:00:52.170 --> 00:00:55.230 align:middle line:84% are on this slide, namely, the fact that, 00:00:55.230 --> 00:00:57.760 align:middle line:84% if you say that these are both grammatical sentences, 00:00:57.760 --> 00:00:59.950 align:middle line:84% but that they mean different things. 00:00:59.950 --> 00:01:02.010 align:middle line:84% The first one, "Susan likes herself," 00:01:02.010 --> 00:01:05.340 align:middle line:84% you could paraphrase as "Susan likes Susan." 00:01:05.340 --> 00:01:08.460 align:middle line:90% "Herself" has to refer to Susan. 00:01:08.460 --> 00:01:13.260 align:middle line:84% And possibly relatedly, "Susan likes her." 00:01:13.260 --> 00:01:15.030 align:middle line:90% You don't know who "her" is. 00:01:15.030 --> 00:01:18.300 align:middle line:84% But you do one thing, which is that it is not Susan. 00:01:18.300 --> 00:01:20.550 align:middle line:84% That's not a general fact about "her." 00:01:20.550 --> 00:01:23.020 align:middle line:84% It's a fact about "her" in this sentence. 00:01:23.020 --> 00:01:24.720 align:middle line:84% So if I said "I like her," then "her" 00:01:24.720 --> 00:01:27.060 align:middle line:84% could be Susan, whoever Susan is. 00:01:27.060 --> 00:01:29.820 align:middle line:84% But if it's "Susan likes her," then that "her" 00:01:29.820 --> 00:01:33.290 align:middle line:84% cannot be Susan, or at least not the same Susan as the subject, 00:01:33.290 --> 00:01:34.620 align:middle line:90% yeah? 00:01:34.620 --> 00:01:39.090 align:middle line:84% Binding theory is meant to capture facts like this. 00:01:39.090 --> 00:01:44.460 align:middle line:84% And these starred sentences incorporate some of the things 00:01:44.460 --> 00:01:46.710 align:middle line:90% that I just said, not only-- 00:01:46.710 --> 00:01:48.852 align:middle line:84% so these little letters, these subscripts 00:01:48.852 --> 00:01:50.310 align:middle line:84% that people put on these words, are 00:01:50.310 --> 00:01:53.880 align:middle line:84% meant to indicate things about what can corefer with what. 00:01:53.880 --> 00:01:58.770 align:middle line:84% So "Susan likes herself," the first sentence, is OK. 00:01:58.770 --> 00:02:00.660 align:middle line:84% And the third sentence is starred. 00:02:00.660 --> 00:02:03.480 align:middle line:84% Because if you say the string "Susan likes herself," 00:02:03.480 --> 00:02:06.630 align:middle line:84% then "Susan" and "herself" have to refer to the same person. 00:02:06.630 --> 00:02:10.139 align:middle line:84% And we indicate that by giving them the same subscript. 00:02:10.139 --> 00:02:14.520 align:middle line:84% So "Susan" sub a likes "herself" sub a, that's OK. 00:02:14.520 --> 00:02:17.970 align:middle line:84% But "Susan" sub a likes "herself" sub b is not OK. 00:02:17.970 --> 00:02:20.433 align:middle line:84% The two subscripts have to be the same. 00:02:20.433 --> 00:02:22.350 align:middle line:84% And with a pronoun, it's the other way around. 00:02:22.350 --> 00:02:23.735 align:middle line:90% They have to be different. 00:02:23.735 --> 00:02:25.103 align:middle line:90% OK so far? 00:02:25.103 --> 00:02:27.270 align:middle line:84% Nobody is sitting there quietly thinking, "No, wait. 00:02:27.270 --> 00:02:30.180 align:middle line:84% I say 'Susan likes her,' and I mean 'Susan likes herself' all 00:02:30.180 --> 00:02:31.150 align:middle line:90% the time. 00:02:31.150 --> 00:02:33.497 align:middle line:84% No wonder nobody ever understands me." 00:02:33.497 --> 00:02:34.830 align:middle line:90% Nobody's having that experience? 00:02:34.830 --> 00:02:37.800 align:middle line:90% OK, good. 00:02:37.800 --> 00:02:41.940 align:middle line:84% So, you know, looking at this without thinking about it 00:02:41.940 --> 00:02:44.790 align:middle line:84% too hard, we could think, OK, maybe pronouns-- 00:02:44.790 --> 00:02:47.550 align:middle line:84% we'll use "pronoun" to refer to words like "her"-- 00:02:47.550 --> 00:02:51.660 align:middle line:84% cannot corefer with anything in the sentence, where "corefer" 00:02:51.660 --> 00:02:54.600 align:middle line:90% means refer to the same person. 00:02:54.600 --> 00:02:59.130 align:middle line:84% And anaphors, which is a name for fancy words like "herself," 00:02:59.130 --> 00:03:00.810 align:middle line:90% must corefer with something. 00:03:00.810 --> 00:03:03.210 align:middle line:84% So "Susan likes herself, herself" 00:03:03.210 --> 00:03:04.800 align:middle line:90% needs to refer back to Susan. 00:03:04.800 --> 00:03:08.100 align:middle line:84% You have to give "herself" something for it 00:03:08.100 --> 00:03:09.090 align:middle line:90% to corefer with. 00:03:09.090 --> 00:03:15.520 align:middle line:90% 00:03:15.520 --> 00:03:17.650 align:middle line:84% And when we did binding theory before, 00:03:17.650 --> 00:03:19.270 align:middle line:90% we already got this far. 00:03:19.270 --> 00:03:21.430 align:middle line:84% A word like "herself" has this special property 00:03:21.430 --> 00:03:25.450 align:middle line:84% that it has to corefer with something else in the sentence. 00:03:25.450 --> 00:03:29.950 align:middle line:84% But interesting fact, it can't just refer with anything. 00:03:29.950 --> 00:03:32.177 align:middle line:84% And this is why there's a theory. 00:03:32.177 --> 00:03:33.760 align:middle line:84% You have to do some work to figure out 00:03:33.760 --> 00:03:35.930 align:middle line:84% which kinds of things it can refer to. 00:03:35.930 --> 00:03:38.680 align:middle line:84% So "Susan likes herself" is fine, 00:03:38.680 --> 00:03:42.380 align:middle line:84% but "Susan's father likes herself," 00:03:42.380 --> 00:03:47.000 align:middle line:84% where "herself" is Susan, is no good. 00:03:47.000 --> 00:03:49.190 align:middle line:84% One hopes that Susan's father likes Susan, 00:03:49.190 --> 00:03:51.020 align:middle line:84% but this is not the way to say it. 00:03:51.020 --> 00:03:53.360 align:middle line:84% So "herself" can't refer back to Susan 00:03:53.360 --> 00:03:55.940 align:middle line:84% in a sentence like this, which knocks out 00:03:55.940 --> 00:03:58.250 align:middle line:84% some of the simplest, easiest theories you might 00:03:58.250 --> 00:04:02.630 align:middle line:84% have had from the first batch of sentences on the earlier slide. 00:04:02.630 --> 00:04:04.340 align:middle line:84% You might have thought, well, you know, 00:04:04.340 --> 00:04:06.110 align:middle line:84% anaphors, they need to refer back to something. 00:04:06.110 --> 00:04:07.610 align:middle line:84% And you might even have thought they 00:04:07.610 --> 00:04:10.250 align:middle line:84% have to refer back to something that's earlier in the sentence. 00:04:10.250 --> 00:04:13.700 align:middle line:84% They're kind of a pointer that points back to something 00:04:13.700 --> 00:04:14.492 align:middle line:90% that's before them. 00:04:14.492 --> 00:04:15.908 align:middle line:84% And what we're seeing is that it's 00:04:15.908 --> 00:04:17.209 align:middle line:90% more complicated than that. 00:04:17.209 --> 00:04:20.400 align:middle line:84% They can't just pick out anything. 00:04:20.400 --> 00:04:22.650 align:middle line:84% There are some restrictions on what they can refer to. 00:04:22.650 --> 00:04:24.232 align:middle line:84% We have to understand what those are. 00:04:24.232 --> 00:04:26.690 align:middle line:84% There's a classic story, which we've already been through-- 00:04:26.690 --> 00:04:28.130 align:middle line:90% I'll just show it to you again-- 00:04:28.130 --> 00:04:31.970 align:middle line:84% about why it's OK to say "Susan must like herself," 00:04:31.970 --> 00:04:36.230 align:middle line:84% but bad to say "Susan's father must like herself." 00:04:36.230 --> 00:04:38.083 align:middle line:84% The classic story goes like this. 00:04:38.083 --> 00:04:39.500 align:middle line:84% What we're trying to do is come up 00:04:39.500 --> 00:04:42.620 align:middle line:84% with a structural relationship between "Susan" 00:04:42.620 --> 00:04:46.220 align:middle line:84% and "herself" that holds in the tree in the left, 00:04:46.220 --> 00:04:48.557 align:middle line:84% but not in the tree in the right. 00:04:48.557 --> 00:04:50.390 align:middle line:84% We're talking about structural relationships 00:04:50.390 --> 00:04:54.500 align:middle line:84% because, well, your first hope, which might have been, 00:04:54.500 --> 00:04:57.980 align:middle line:84% yeah, "herself" has to refer to something that's been mentioned 00:04:57.980 --> 00:05:02.220 align:middle line:84% before, the way you might have hoped language was designed, 00:05:02.220 --> 00:05:03.740 align:middle line:84% there are no languages like that. 00:05:03.740 --> 00:05:05.510 align:middle line:84% There are, instead, languages like this 00:05:05.510 --> 00:05:07.760 align:middle line:84% that apparently care about trees when 00:05:07.760 --> 00:05:09.230 align:middle line:84% they're deciding which things can 00:05:09.230 --> 00:05:12.650 align:middle line:90% refer to which other things. 00:05:12.650 --> 00:05:14.150 align:middle line:84% And so we're going to try to develop 00:05:14.150 --> 00:05:17.330 align:middle line:84% a structural distinction between the relationship 00:05:17.330 --> 00:05:20.047 align:middle line:84% between "Susan" and "herself" in the tree on the left 00:05:20.047 --> 00:05:22.130 align:middle line:84% and the relationship between "Susan" and "herself" 00:05:22.130 --> 00:05:23.590 align:middle line:90% in the tree on the right. 00:05:23.590 --> 00:05:25.340 align:middle line:84% As I said, we've talked about this before. 00:05:25.340 --> 00:05:27.280 align:middle line:90% We're just reviewing it. 00:05:27.280 --> 00:05:30.430 align:middle line:84% The story that people have classically come up with 00:05:30.430 --> 00:05:33.940 align:middle line:84% makes use of the fact that Susan is more deeply 00:05:33.940 --> 00:05:38.090 align:middle line:84% buried in the tree on the right than she is on the left. 00:05:38.090 --> 00:05:40.405 align:middle line:84% So there's a relation called c-command. 00:05:40.405 --> 00:05:43.540 align:middle line:90% 00:05:43.540 --> 00:05:47.605 align:middle line:84% It's a relation that can hold between two nodes in a tree. 00:05:47.605 --> 00:05:52.600 align:middle line:84% It says alpha c-commands beta if every node that dominates alpha 00:05:52.600 --> 00:05:54.790 align:middle line:90% dominates beta. 00:05:54.790 --> 00:05:58.300 align:middle line:84% So you c-command basically the thing that you were merged with 00:05:58.300 --> 00:06:01.450 align:middle line:84% and everything that that dominates. 00:06:01.450 --> 00:06:07.810 align:middle line:84% So in this tree, the NP "Susan" c-commands "herself." 00:06:07.810 --> 00:06:09.340 align:middle line:84% The way we find that out is we ask, 00:06:09.340 --> 00:06:12.190 align:middle line:84% what are the nodes that dominate the noun phrase "Susan"? 00:06:12.190 --> 00:06:15.140 align:middle line:84% And there's only one, that TP that I've put up there. 00:06:15.140 --> 00:06:17.170 align:middle line:84% And that TP also dominates "herself," 00:06:17.170 --> 00:06:20.470 align:middle line:84% and so "Susan" c-commands "herself." 00:06:20.470 --> 00:06:23.810 align:middle line:84% Does "herself" c-command "Susan"? 00:06:23.810 --> 00:06:24.310 align:middle line:90% No. 00:06:24.310 --> 00:06:26.685 align:middle line:84% Yeah, you're all shaking your heads appropriately, right? 00:06:26.685 --> 00:06:28.350 align:middle line:84% And the way to find that out is to ask, 00:06:28.350 --> 00:06:30.920 align:middle line:84% what are the nodes that dominate herself? 00:06:30.920 --> 00:06:33.590 align:middle line:84% And they include that TP, but somebody name another one 00:06:33.590 --> 00:06:35.380 align:middle line:90% that doesn't dominate "Susan." 00:06:35.380 --> 00:06:36.017 align:middle line:90% AUDIENCE: VP. 00:06:36.017 --> 00:06:37.100 align:middle line:90% NORVIN RICHARDS: VP, yeah. 00:06:37.100 --> 00:06:38.000 align:middle line:90% Or T bar, yeah? 00:06:38.000 --> 00:06:41.465 align:middle line:84% So there are at least those two which dominate "herself," 00:06:41.465 --> 00:06:42.590 align:middle line:90% but don't dominate "Susan." 00:06:42.590 --> 00:06:44.390 align:middle line:84% So "Susan" c-commands "herself," but "herself" 00:06:44.390 --> 00:06:45.590 align:middle line:90% doesn't c-command "Susan." 00:06:45.590 --> 00:06:49.460 align:middle line:84% And what we're seeing is that maybe an anaphor 00:06:49.460 --> 00:06:52.110 align:middle line:84% needs to refer to something that c-commands it. 00:06:52.110 --> 00:06:54.860 align:middle line:84% So in this sentence, "Susan" must like herself," 00:06:54.860 --> 00:06:56.810 align:middle line:90% "herself" is referring to Susan. 00:06:56.810 --> 00:06:59.240 align:middle line:84% And "Susan" c-commands "herself," as opposed 00:06:59.240 --> 00:07:02.700 align:middle line:84% to the other tree, "Susan's father must like herself," 00:07:02.700 --> 00:07:03.855 align:middle line:90% which is no good. 00:07:03.855 --> 00:07:04.730 align:middle line:90% What's wrong with it? 00:07:04.730 --> 00:07:07.430 align:middle line:84% Well, what's wrong with it is, on this story, 00:07:07.430 --> 00:07:10.340 align:middle line:84% there are multiple nodes dominating "Susan." 00:07:10.340 --> 00:07:13.070 align:middle line:90% There's TP, but also NP and DP. 00:07:13.070 --> 00:07:15.500 align:middle line:84% And some of those don't dominate "herself." 00:07:15.500 --> 00:07:18.920 align:middle line:84% So "Susan" doesn't c-command "herself." 00:07:18.920 --> 00:07:21.925 align:middle line:84% Binding theory makes use of this. 00:07:21.925 --> 00:07:23.300 align:middle line:84% Let me just call your attention-- 00:07:23.300 --> 00:07:26.240 align:middle line:84% maybe your attention is already firmly fixed on this-- 00:07:26.240 --> 00:07:28.190 align:middle line:90% to how weird this is. 00:07:28.190 --> 00:07:33.710 align:middle line:84% I just alluded to this, but let me just say it again. 00:07:33.710 --> 00:07:36.110 align:middle line:84% I'm showing you a bunch of facts about English. 00:07:36.110 --> 00:07:38.390 align:middle line:84% And different languages do have different anaphors 00:07:38.390 --> 00:07:40.230 align:middle line:90% that behave in different ways. 00:07:40.230 --> 00:07:43.350 align:middle line:84% But here's something no language ever does, as far as I know. 00:07:43.350 --> 00:07:45.020 align:middle line:84% There are languages in the world where 00:07:45.020 --> 00:07:49.370 align:middle line:84% an anaphor can refer to anything that was said earlier. 00:07:49.370 --> 00:07:53.432 align:middle line:84% There are no languages in which both "Susan likes herself" 00:07:53.432 --> 00:07:54.890 align:middle line:84% and "Susan's father likes herself," 00:07:54.890 --> 00:07:56.330 align:middle line:90% in which they're both OK. 00:07:56.330 --> 00:07:58.130 align:middle line:84% There just aren't any languages like that 00:07:58.130 --> 00:08:01.820 align:middle line:84% in which the important thing is does the noun precede 00:08:01.820 --> 00:08:06.440 align:middle line:84% the anaphor, which is, so first of all, interesting. 00:08:06.440 --> 00:08:08.090 align:middle line:84% There aren't any languages like that. 00:08:08.090 --> 00:08:12.483 align:middle line:84% But second of all, weirdly, surprisingly, interesting, 00:08:12.483 --> 00:08:14.150 align:middle line:84% let's get ourselves in the frame of mind 00:08:14.150 --> 00:08:16.100 align:middle line:84% where we can be impressed by a fact like that. 00:08:16.100 --> 00:08:19.760 align:middle line:84% Because, look, imagine what you were like when 00:08:19.760 --> 00:08:20.960 align:middle line:90% you were a baby, right? 00:08:20.960 --> 00:08:21.560 align:middle line:90% Here you are. 00:08:21.560 --> 00:08:22.850 align:middle line:90% You've just been born. 00:08:22.850 --> 00:08:24.830 align:middle line:84% You're hearing people speak, right? 00:08:24.830 --> 00:08:27.040 align:middle line:84% They're saying sentences in this language that's 00:08:27.040 --> 00:08:28.415 align:middle line:84% going to be your native language, 00:08:28.415 --> 00:08:29.480 align:middle line:90% and you're trying to learn it. 00:08:29.480 --> 00:08:31.340 align:middle line:84% You're trying to figure things out about it. 00:08:31.340 --> 00:08:34.010 align:middle line:84% What's the one thing that you know? 00:08:34.010 --> 00:08:35.360 align:middle line:90% You hear people talking. 00:08:35.360 --> 00:08:37.309 align:middle line:84% And they're saying things like-- here, 00:08:37.309 --> 00:08:39.539 align:middle line:84% I'll put a sentence on the board in a language 00:08:39.539 --> 00:08:40.789 align:middle line:90% that I hope none of you speak. 00:08:40.789 --> 00:08:48.530 align:middle line:90% 00:08:48.530 --> 00:08:50.510 align:middle line:90% So here's a sentence. 00:08:50.510 --> 00:08:55.910 align:middle line:84% You hear one of your parents saying, [NON-ENGLISH SPEECH].. 00:08:55.910 --> 00:08:57.860 align:middle line:84% Maybe they write it down and show it to you. 00:08:57.860 --> 00:08:59.485 align:middle line:84% So you know, for some reason, that it's 00:08:59.485 --> 00:09:01.970 align:middle line:90% 1, 2, 3, 4, 5 words long. 00:09:01.970 --> 00:09:05.893 align:middle line:84% What else do you know about that sentence? 00:09:05.893 --> 00:09:07.560 align:middle line:84% Do you know which words in that sentence 00:09:07.560 --> 00:09:09.940 align:middle line:90% c-command which other words? 00:09:09.940 --> 00:09:11.400 align:middle line:90% Good lord, no. 00:09:11.400 --> 00:09:12.583 align:middle line:90% Yeah? 00:09:12.583 --> 00:09:14.250 align:middle line:84% Do you know which words in that sentence 00:09:14.250 --> 00:09:16.490 align:middle line:90% precede which other words? 00:09:16.490 --> 00:09:17.600 align:middle line:90% Yeah. 00:09:17.600 --> 00:09:20.975 align:middle line:84% So you know that [NON-ENGLISH] precedes [NON-ENGLISH].. 00:09:20.975 --> 00:09:22.405 align:middle line:90% You know that. 00:09:22.405 --> 00:09:24.530 align:middle line:84% But you don't know whether [NON-ENGLISH] c-commands 00:09:24.530 --> 00:09:25.890 align:middle line:90% [NON-ENGLISH] or not. 00:09:25.890 --> 00:09:28.550 align:middle line:84% So the one bit of information that you have immediately 00:09:28.550 --> 00:09:32.370 align:middle line:84% from the get-go, which words come before which other words, 00:09:32.370 --> 00:09:34.610 align:middle line:90% you ignore. 00:09:34.610 --> 00:09:36.920 align:middle line:84% So languages don't ever make use of that when 00:09:36.920 --> 00:09:38.490 align:middle line:90% they're doing binding theory. 00:09:38.490 --> 00:09:44.000 align:middle line:84% So you would have thought this one sort of easy to access 00:09:44.000 --> 00:09:45.410 align:middle line:84% bit of information that all of us 00:09:45.410 --> 00:09:48.365 align:middle line:84% have from day one, no matter, you know-- 00:09:48.365 --> 00:09:50.240 align:middle line:84% don't know anything else about your language. 00:09:50.240 --> 00:09:53.090 align:middle line:84% You at least know which things come before which other things. 00:09:53.090 --> 00:09:55.688 align:middle line:84% Binding theory never, ever cares about that. 00:09:55.688 --> 00:09:57.230 align:middle line:84% It only cares about c-commands, which 00:09:57.230 --> 00:09:59.120 align:middle line:84% is this thing that you have to know something 00:09:59.120 --> 00:10:00.560 align:middle line:90% in order to understand. 00:10:00.560 --> 00:10:03.550 align:middle line:90% 00:10:03.550 --> 00:10:05.890 align:middle line:84% It's as though our brains are designed 00:10:05.890 --> 00:10:08.980 align:middle line:84% to build language in some ways and not others 00:10:08.980 --> 00:10:14.620 align:middle line:84% and to ignore certain sources of data 00:10:14.620 --> 00:10:16.148 align:middle line:90% which are very, very fruitful. 00:10:16.148 --> 00:10:18.190 align:middle line:84% You have lots and lots of data about which things 00:10:18.190 --> 00:10:20.868 align:middle line:84% precede which other things, but you don't care. 00:10:20.868 --> 00:10:21.910 align:middle line:90% It's kind of interesting. 00:10:21.910 --> 00:10:23.770 align:middle line:90% Yeah. 00:10:23.770 --> 00:10:26.590 align:middle line:84% AUDIENCE: If, for some reason, we 00:10:26.590 --> 00:10:29.823 align:middle line:84% did care about the order of the words that we're saying, 00:10:29.823 --> 00:10:30.990 align:middle line:90% would that create ambiguity? 00:10:30.990 --> 00:10:35.790 align:middle line:84% Because some languages like, German, half of it 00:10:35.790 --> 00:10:38.108 align:middle line:84% is head initial and half is head final. 00:10:38.108 --> 00:10:40.035 align:middle line:90% And so you'd be confused. 00:10:40.035 --> 00:10:42.410 align:middle line:84% NORVIN RICHARDS: So which things proceed and follow which 00:10:42.410 --> 00:10:42.990 align:middle line:90% other things? 00:10:42.990 --> 00:10:44.990 align:middle line:84% I mean, there are ambiguities in binding theory. 00:10:44.990 --> 00:10:47.032 align:middle line:84% We've been talking about simple sentences so far, 00:10:47.032 --> 00:10:50.090 align:middle line:84% but you can say things like-- if anybody is wondering, 00:10:50.090 --> 00:10:51.320 align:middle line:90% this [NON-ENGLISH]. 00:10:51.320 --> 00:10:59.070 align:middle line:84% It means "Is drinking the man the water," 00:10:59.070 --> 00:11:01.130 align:middle line:84% so "The man is drinking the water." 00:11:01.130 --> 00:11:03.218 align:middle line:84% Sorry, I just told you that so you wouldn't 00:11:03.218 --> 00:11:04.260 align:middle line:90% have to worry about that. 00:11:04.260 --> 00:11:05.630 align:middle line:84% There are ambiguities in binding theory. 00:11:05.630 --> 00:11:07.250 align:middle line:84% I've been talking about sentences where there aren't 00:11:07.250 --> 00:11:08.708 align:middle line:84% ambiguities, but you can say things 00:11:08.708 --> 00:11:19.210 align:middle line:84% like "Susan told Stacy about herself," 00:11:19.210 --> 00:11:23.260 align:middle line:84% where probably the easiest reading for that 00:11:23.260 --> 00:11:26.560 align:middle line:84% is that Susan told Stacy about Susan. 00:11:26.560 --> 00:11:32.025 align:middle line:84% But if Stacy is unclear about herself, 00:11:32.025 --> 00:11:34.150 align:middle line:84% there's something Stacy doesn't know about herself, 00:11:34.150 --> 00:11:35.800 align:middle line:84% Susan could tell Stacy about herself 00:11:35.800 --> 00:11:39.010 align:middle line:84% and have that mean Susan told Stacy about Stacy, I think. 00:11:39.010 --> 00:11:40.220 align:middle line:90% Am I making that up? 00:11:40.220 --> 00:11:41.110 align:middle line:90% I think that's true. 00:11:41.110 --> 00:11:42.730 align:middle line:90% This is just ambiguous. 00:11:42.730 --> 00:11:43.700 align:middle line:90% So here's a sentence. 00:11:43.700 --> 00:11:46.570 align:middle line:84% So we're OK with ambiguity usually. 00:11:46.570 --> 00:11:48.142 align:middle line:84% There are plenty of places where, 00:11:48.142 --> 00:11:49.600 align:middle line:84% if we were to draw a tree for this, 00:11:49.600 --> 00:11:51.700 align:middle line:84% we'd want to draw a tree where "Susan" and "Stacy" 00:11:51.700 --> 00:11:53.500 align:middle line:90% both c-command "herself." 00:11:53.500 --> 00:11:56.050 align:middle line:84% And so it has a choice of which kind of thing 00:11:56.050 --> 00:11:57.310 align:middle line:90% it wants to refer to. 00:11:57.310 --> 00:11:59.558 align:middle line:84% All it cares about is which thing c-commands it, 00:11:59.558 --> 00:12:01.600 align:middle line:84% but it doesn't care about which thing precedes it 00:12:01.600 --> 00:12:03.250 align:middle line:90% in any language on Earth. 00:12:03.250 --> 00:12:05.170 align:middle line:90% It was kind of interesting. 00:12:05.170 --> 00:12:06.040 align:middle line:90% Yeah, go. 00:12:06.040 --> 00:12:08.540 align:middle line:84% AUDIENCE: I was also thinking about you could have something 00:12:08.540 --> 00:12:12.450 align:middle line:84% like "Carrie likes Mary herself." 00:12:12.450 --> 00:12:16.710 align:middle line:84% Maybe Mary played the evil character on TV and people 00:12:16.710 --> 00:12:18.210 align:middle line:84% don't like-- she likes Mary herself. 00:12:18.210 --> 00:12:19.460 align:middle line:90% NORVIN RICHARDS: Right, right. 00:12:19.460 --> 00:12:22.230 align:middle line:84% AUDIENCE: But it's like, all the other people, they hate Mary. 00:12:22.230 --> 00:12:24.573 align:middle line:90% But she, herself, she likes-- 00:12:24.573 --> 00:12:25.740 align:middle line:90% NORVIN RICHARDS: She-- yeah. 00:12:25.740 --> 00:12:28.265 align:middle line:84% You're raising-- actually, I'm glad you asked that question 00:12:28.265 --> 00:12:30.640 align:middle line:84% because it raises a good point, which I should have made. 00:12:30.640 --> 00:12:34.920 align:middle line:84% And so thank you for raising it, which is that languages use 00:12:34.920 --> 00:12:38.910 align:middle line:84% "herself," "himself," "myself," all these "-self" 00:12:38.910 --> 00:12:42.390 align:middle line:84% kinds of doohickeys, for a bunch of different things. 00:12:42.390 --> 00:12:46.290 align:middle line:84% We're going to be concentrating on the "herself" that 00:12:46.290 --> 00:12:48.960 align:middle line:84% sits where a noun phrase would normally sit 00:12:48.960 --> 00:12:52.590 align:middle line:84% and refers back to another noun phrase. 00:12:52.590 --> 00:12:54.780 align:middle line:84% But there is this other, sometimes called, 00:12:54.780 --> 00:12:57.630 align:middle line:90% an intensive use of reflexives. 00:12:57.630 --> 00:12:59.220 align:middle line:90% And I can't spell today-- 00:12:59.220 --> 00:13:13.020 align:middle line:84% intensive, where you say things like "John, himself, fixed 00:13:13.020 --> 00:13:16.980 align:middle line:84% the computer," or, for that matter, 00:13:16.980 --> 00:13:20.380 align:middle line:84% "John fixed the computer himself," 00:13:20.380 --> 00:13:27.270 align:middle line:84% where that means something not so different from "John 00:13:27.270 --> 00:13:29.010 align:middle line:90% fixed the computer." 00:13:29.010 --> 00:13:31.080 align:middle line:84% It means something like "John fixed the computer, 00:13:31.080 --> 00:13:33.990 align:middle line:90% and no one else did." 00:13:33.990 --> 00:13:37.060 align:middle line:84% So "Susan's father must like herself." 00:13:37.060 --> 00:13:40.230 align:middle line:84% We were talking about things that "herself" can refer to. 00:13:40.230 --> 00:13:42.168 align:middle line:84% Or "Susan told Stacy about herself." 00:13:42.168 --> 00:13:44.460 align:middle line:84% We're talking about things that "herself" can refer to. 00:13:44.460 --> 00:13:46.860 align:middle line:84% This would be an OK sentence if you changed "herself" 00:13:46.860 --> 00:13:47.400 align:middle line:90% to something else. 00:13:47.400 --> 00:13:48.608 align:middle line:90% It would mean something else. 00:13:48.608 --> 00:13:52.080 align:middle line:84% But "Susan told Stacy about Mary," 00:13:52.080 --> 00:13:53.610 align:middle line:90% that's also a sentence, right? 00:13:53.610 --> 00:13:55.390 align:middle line:84% So "herself" is just a noun phrase, 00:13:55.390 --> 00:13:57.060 align:middle line:84% noun phrase with special properties, 00:13:57.060 --> 00:13:59.370 align:middle line:84% that's sitting where noun phrases go. 00:13:59.370 --> 00:14:02.500 align:middle line:84% If you replaced "himself" here with something else, 00:14:02.500 --> 00:14:05.940 align:middle line:84% so if it were "John, Mary fixed the computer," 00:14:05.940 --> 00:14:09.450 align:middle line:84% you would no longer be speaking English. 00:14:09.450 --> 00:14:12.345 align:middle line:84% So this is a different construction, the construction 00:14:12.345 --> 00:14:14.220 align:middle line:84% that you're raising, a really interesting one 00:14:14.220 --> 00:14:17.790 align:middle line:84% which I'm now going to ignore for the rest of today. 00:14:17.790 --> 00:14:20.940 align:middle line:90% But that's what this is. 00:14:20.940 --> 00:14:21.823 align:middle line:90% Yeah. 00:14:21.823 --> 00:14:23.490 align:middle line:84% AUDIENCE: Well, I'm bringing this again. 00:14:23.490 --> 00:14:24.330 align:middle line:90% NORVIN RICHARDS: Oh, no. 00:14:24.330 --> 00:14:25.090 align:middle line:90% I'm ignoring you. 00:14:25.090 --> 00:14:25.590 align:middle line:90% No, sorry. 00:14:25.590 --> 00:14:26.480 align:middle line:90% Go ahead. 00:14:26.480 --> 00:14:27.480 align:middle line:84% AUDIENCE: Is that a type of movement? 00:14:27.480 --> 00:14:28.020 align:middle line:90% NORVIN RICHARDS: Sorry? 00:14:28.020 --> 00:14:29.970 align:middle line:84% AUDIENCE: Is that another example of movement? 00:14:29.970 --> 00:14:30.510 align:middle line:90% Or-- 00:14:30.510 --> 00:14:31.050 align:middle line:90% NORVIN RICHARDS: Which one? 00:14:31.050 --> 00:14:33.258 align:middle line:84% Oh, the fact that himself can be either here or here? 00:14:33.258 --> 00:14:34.412 align:middle line:90% AUDIENCE: Yeah. 00:14:34.412 --> 00:14:35.370 align:middle line:90% NORVIN RICHARDS: Maybe. 00:14:35.370 --> 00:14:39.640 align:middle line:90% 00:14:39.640 --> 00:14:43.490 align:middle line:90% Ah, we have to be careful. 00:14:43.490 --> 00:14:47.770 align:middle line:84% There are various kinds of-- so himself, here, it's 00:14:47.770 --> 00:14:49.570 align:middle line:90% kind of like an adverb, right? 00:14:49.570 --> 00:14:53.140 align:middle line:84% It's modifying the way in which he fixed the computer. 00:14:53.140 --> 00:14:56.082 align:middle line:84% And just as you can see things like "John quickly"-- 00:14:56.082 --> 00:14:57.790 align:middle line:84% I don't know why I switched to John here. 00:14:57.790 --> 00:14:59.170 align:middle line:90% Let's make Susan do this, too. 00:14:59.170 --> 00:15:03.220 align:middle line:90% 00:15:03.220 --> 00:15:06.010 align:middle line:84% Just as you can say Susan quickly fixed the computer, 00:15:06.010 --> 00:15:09.850 align:middle line:84% or Susan fixed the computer quickly, 00:15:09.850 --> 00:15:11.860 align:middle line:90% you could do that with movement. 00:15:11.860 --> 00:15:12.965 align:middle line:90% Or you could decide-- 00:15:12.965 --> 00:15:14.590 align:middle line:84% we're learning something about adverbs, 00:15:14.590 --> 00:15:17.140 align:middle line:84% that adverbs, because it's an adjunct maybe, 00:15:17.140 --> 00:15:19.270 align:middle line:84% there are more options about where it can go. 00:15:19.270 --> 00:15:22.750 align:middle line:84% The rules about where exactly it goes are not as strict. 00:15:22.750 --> 00:15:25.090 align:middle line:84% The clean arguments for movement are 00:15:25.090 --> 00:15:28.300 align:middle line:84% cases where we know for a fact that something should be here, 00:15:28.300 --> 00:15:31.150 align:middle line:90% and yet it's over here. 00:15:31.150 --> 00:15:33.910 align:middle line:90% When we do "What did he devour?" 00:15:33.910 --> 00:15:35.380 align:middle line:90% that's what we're doing there. 00:15:35.380 --> 00:15:36.130 align:middle line:90% Yes? 00:15:36.130 --> 00:15:37.750 align:middle line:84% AUDIENCE: But doesn't the "you" in that sentence kind 00:15:37.750 --> 00:15:40.058 align:middle line:84% of change depending on where you put the "herself"? 00:15:40.058 --> 00:15:41.350 align:middle line:90% NORVIN RICHARDS: Oh, sometimes. 00:15:41.350 --> 00:15:42.227 align:middle line:90% Yeah. 00:15:42.227 --> 00:15:43.510 align:middle line:84% AUDIENCE: The first one, you could have "herself" 00:15:43.510 --> 00:15:45.827 align:middle line:84% after Susan if you're implying that it's surprising 00:15:45.827 --> 00:15:48.160 align:middle line:84% that Susan, herself, was the one who fixed the computer. 00:15:48.160 --> 00:15:48.760 align:middle line:90% NORVIN RICHARDS: Yeah. 00:15:48.760 --> 00:15:50.650 align:middle line:84% AUDIENCE: Or as you put it at the end, it's you just 00:15:50.650 --> 00:15:51.700 align:middle line:90% say that it was Susan that-- 00:15:51.700 --> 00:15:52.660 align:middle line:84% NORVIN RICHARDS: Yeah, so these might 00:15:52.660 --> 00:15:53.870 align:middle line:90% be different kinds of adverbs. 00:15:53.870 --> 00:15:54.700 align:middle line:90% That's a very good point. 00:15:54.700 --> 00:15:55.970 align:middle line:90% Yeah, that's a very nice point. 00:15:55.970 --> 00:15:58.012 align:middle line:84% So if I say "Susan, herself, fixed the computer," 00:15:58.012 --> 00:15:59.853 align:middle line:84% we mean Susan is the CEO of the company. 00:15:59.853 --> 00:16:02.270 align:middle line:84% You would have expected her to hire someone else to do it, 00:16:02.270 --> 00:16:04.540 align:middle line:84% but she did it herself, whereas "Susan 00:16:04.540 --> 00:16:08.290 align:middle line:84% fixed the computer herself," it doesn't mean necessarily 00:16:08.290 --> 00:16:09.860 align:middle line:84% that it's surprising that she did it. 00:16:09.860 --> 00:16:12.340 align:middle line:84% It just means that she did it and no one else did. 00:16:12.340 --> 00:16:13.090 align:middle line:90% Yeah. 00:16:13.090 --> 00:16:15.670 align:middle line:84% So they're related meanings, but separate. 00:16:15.670 --> 00:16:18.730 align:middle line:84% So we may want to distinguish two different types 00:16:18.730 --> 00:16:19.930 align:middle line:90% of intensive modification. 00:16:19.930 --> 00:16:22.267 align:middle line:84% You're absolutely right, nice point. 00:16:22.267 --> 00:16:23.850 align:middle line:84% Other nice points people wish to make? 00:16:23.850 --> 00:16:24.350 align:middle line:90% Yes? 00:16:24.350 --> 00:16:27.428 align:middle line:84% AUDIENCE: So to clarify, the reason this tree is not valid 00:16:27.428 --> 00:16:32.220 align:middle line:84% [INAUDIBLE] is that "herself" is looking for something to bind 00:16:32.220 --> 00:16:32.720 align:middle line:90% to. 00:16:32.720 --> 00:16:33.637 align:middle line:90% NORVIN RICHARDS: Yeah. 00:16:33.637 --> 00:16:37.190 align:middle line:84% AUDIENCE: And so it needs to be binded [INAUDIBLE] binds to NP. 00:16:37.190 --> 00:16:39.820 align:middle line:84% It binds "Susan," and it binds "Susan's father." 00:16:39.820 --> 00:16:41.410 align:middle line:84% And it can't bind to "Susan" because 00:16:41.410 --> 00:16:42.742 align:middle line:90% of the c-command constraint. 00:16:42.742 --> 00:16:45.747 align:middle line:84% It can't bind to "Susan's father" because "father"-- 00:16:45.747 --> 00:16:47.330 align:middle line:84% NORVIN RICHARDS: Because wrong gender. 00:16:47.330 --> 00:16:47.570 align:middle line:90% AUDIENCE: Right. 00:16:47.570 --> 00:16:48.050 align:middle line:90% NORVIN RICHARDS: Yup. 00:16:48.050 --> 00:16:48.770 align:middle line:90% AUDIENCE: [INAUDIBLE] himself. 00:16:48.770 --> 00:16:49.645 align:middle line:90% NORVIN RICHARDS: Yup. 00:16:49.645 --> 00:16:51.740 align:middle line:90% Yeah, that's it exactly. 00:16:51.740 --> 00:16:53.690 align:middle line:90% Yeah, so the idea is-- 00:16:53.690 --> 00:16:56.390 align:middle line:84% so we weren't asking why can't it refer to Susan's father. 00:16:56.390 --> 00:16:58.098 align:middle line:84% You just asked that question, and I think 00:16:58.098 --> 00:17:00.850 align:middle line:90% you answered it perfectly. 00:17:00.850 --> 00:17:02.600 align:middle line:84% "Herself" can't refer to "Susan's father," 00:17:02.600 --> 00:17:06.170 align:middle line:84% which does c-command "herself," because Susan's father is male. 00:17:06.170 --> 00:17:06.710 align:middle line:90% Wake up. 00:17:06.710 --> 00:17:09.380 align:middle line:90% 00:17:09.380 --> 00:17:11.329 align:middle line:84% And it can't refer to "Susan" because "Susan" 00:17:11.329 --> 00:17:14.995 align:middle line:90% doesn't c-command it. 00:17:14.995 --> 00:17:15.810 align:middle line:90% Is it back? 00:17:15.810 --> 00:17:15.950 align:middle line:90% Good. 00:17:15.950 --> 00:17:16.460 align:middle line:90% Yes. 00:17:16.460 --> 00:17:17.320 align:middle line:90% Yeah. 00:17:17.320 --> 00:17:19.740 align:middle line:84% AUDIENCE: So then to clarify, if the sentence was, 00:17:19.740 --> 00:17:21.750 align:middle line:84% "Susan's father must like himself," 00:17:21.750 --> 00:17:23.609 align:middle line:84% that's fine because "Susan's father" 00:17:23.609 --> 00:17:24.962 align:middle line:90% is the NP that's c-commanding. 00:17:24.962 --> 00:17:25.920 align:middle line:90% NORVIN RICHARDS: Right. 00:17:25.920 --> 00:17:27.660 align:middle line:84% And that's my intuition about the sentence. 00:17:27.660 --> 00:17:28.243 align:middle line:90% Is that yours? 00:17:28.243 --> 00:17:29.760 align:middle line:90% And that would be OK. 00:17:29.760 --> 00:17:34.320 align:middle line:84% It means Susan's father has a healthy self-regard. 00:17:34.320 --> 00:17:35.550 align:middle line:90% Yeah? 00:17:35.550 --> 00:17:36.360 align:middle line:90% OK, cool. 00:17:36.360 --> 00:17:39.600 align:middle line:84% So binding theory, I introduced you to binding theory 00:17:39.600 --> 00:17:41.220 align:middle line:90% as a test for structure, right? 00:17:41.220 --> 00:17:45.450 align:middle line:84% So once we are willing to trust binding theory 00:17:45.450 --> 00:17:47.580 align:middle line:84% at least a little ways, we can convince ourselves 00:17:47.580 --> 00:17:49.680 align:middle line:84% that we have a new way of finding out 00:17:49.680 --> 00:17:50.982 align:middle line:90% things about structure. 00:17:50.982 --> 00:17:52.440 align:middle line:84% We know we have the right structure 00:17:52.440 --> 00:17:54.898 align:middle line:84% if we have the right c-command relations to get the binding 00:17:54.898 --> 00:17:57.400 align:middle line:90% relations that we seem to find. 00:17:57.400 --> 00:18:00.300 align:middle line:84% So for example, when we're thinking about sentences 00:18:00.300 --> 00:18:03.122 align:middle line:84% like this, "Susan told Stacy about herself," 00:18:03.122 --> 00:18:05.580 align:middle line:84% we're going to need it to have a structure in which "Susan" 00:18:05.580 --> 00:18:07.680 align:middle line:84% and "Stacy" both c-command herself. 00:18:07.680 --> 00:18:10.200 align:middle line:84% And so when we were talking before about complicated 00:18:10.200 --> 00:18:12.795 align:middle line:84% verb phrases, binding theory is the kind of thing 00:18:12.795 --> 00:18:14.670 align:middle line:84% that people use as a probe into the structure 00:18:14.670 --> 00:18:16.910 align:middle line:90% of complicated verb phrases. 00:18:16.910 --> 00:18:19.353 align:middle line:90% OK. 00:18:19.353 --> 00:18:21.020 align:middle line:84% All right, so one part of binding theory 00:18:21.020 --> 00:18:23.240 align:middle line:84% is going to say anaphors, these are words 00:18:23.240 --> 00:18:26.210 align:middle line:90% like "herself" or "myself"-- 00:18:26.210 --> 00:18:29.450 align:middle line:84% in English, they all end with "-self." 00:18:29.450 --> 00:18:32.330 align:middle line:84% They need to be c-commanded by something 00:18:32.330 --> 00:18:34.620 align:middle line:90% that corefers with them. 00:18:34.620 --> 00:18:36.582 align:middle line:90% And then we have a short name. 00:18:36.582 --> 00:18:38.040 align:middle line:84% I hope I do this on the next slide. 00:18:38.040 --> 00:18:39.207 align:middle line:90% I'm not sure if I do or not. 00:18:39.207 --> 00:18:42.650 align:middle line:84% We have a short name for to c-command something 00:18:42.650 --> 00:18:45.750 align:middle line:84% that you corefer with, we call that bind. 00:18:45.750 --> 00:18:51.110 align:middle line:84% So we say that, in "Susan told Stacy about herself," 00:18:51.110 --> 00:18:53.060 align:middle line:90% "herself" needs to be bound. 00:18:53.060 --> 00:18:55.460 align:middle line:84% And there are two noun phrases that could bind it, 00:18:55.460 --> 00:18:57.367 align:middle line:90% "Susan" and "Stacy." 00:18:57.367 --> 00:18:59.450 align:middle line:84% And so we need a structure for this in which those 00:18:59.450 --> 00:19:01.850 align:middle line:84% both c-command herself, and either of them 00:19:01.850 --> 00:19:03.830 align:middle line:90% could corefer with herself. 00:19:03.830 --> 00:19:05.570 align:middle line:90% Yeah? 00:19:05.570 --> 00:19:09.290 align:middle line:84% AUDIENCE: Do all anaphors necessarily end in "-self," 00:19:09.290 --> 00:19:12.590 align:middle line:84% or are there cases where pronouns act as anaphors? 00:19:12.590 --> 00:19:15.740 align:middle line:84% So for example, "John said he went to the store," 00:19:15.740 --> 00:19:17.280 align:middle line:90% where "he" is referring to John. 00:19:17.280 --> 00:19:19.280 align:middle line:84% NORVIN RICHARDS: Yeah, that's a very nice point. 00:19:19.280 --> 00:19:21.960 align:middle line:84% So, so far, you may have noticed all of my examples 00:19:21.960 --> 00:19:24.510 align:middle line:90% have only been one clause long. 00:19:24.510 --> 00:19:28.050 align:middle line:84% And we're going to make things more complicated 00:19:28.050 --> 00:19:30.550 align:middle line:90% in just a second. 00:19:30.550 --> 00:19:33.405 align:middle line:84% But here, just to get started on making things more complicated, 00:19:33.405 --> 00:19:35.530 align:middle line:84% we've been talking about the fact that you can say, 00:19:35.530 --> 00:19:43.110 align:middle line:90% "Susan must like herself." 00:19:43.110 --> 00:19:45.100 align:middle line:84% If we make this an embedded clause, 00:19:45.100 --> 00:19:51.870 align:middle line:84% so if we say "Mary thinks Susan must like herself," well, 00:19:51.870 --> 00:19:54.520 align:middle line:90% "herself" can refer to Susan. 00:19:54.520 --> 00:19:57.460 align:middle line:90% But it can't refer to Mary. 00:19:57.460 --> 00:19:59.950 align:middle line:84% So "Mary thinks Susan must like herself." 00:19:59.950 --> 00:20:02.173 align:middle line:84% And that's not going to be because "Mary" 00:20:02.173 --> 00:20:03.340 align:middle line:90% doesn't c-command "herself." 00:20:03.340 --> 00:20:05.740 align:middle line:84% It does in the kinds of trees that we're drawing. 00:20:05.740 --> 00:20:08.450 align:middle line:84% So you are finding a complication, 00:20:08.450 --> 00:20:10.870 align:middle line:84% which I was planning to hide for a few more slides. 00:20:10.870 --> 00:20:13.540 align:middle line:84% There's an additional requirement on anaphors. 00:20:13.540 --> 00:20:16.673 align:middle line:84% They don't just need to be c-commanded by something 00:20:16.673 --> 00:20:17.590 align:middle line:90% that corefers to them. 00:20:17.590 --> 00:20:19.000 align:middle line:84% They don't just need to be bound. 00:20:19.000 --> 00:20:20.417 align:middle line:84% They need to be bound by something 00:20:20.417 --> 00:20:22.690 align:middle line:90% that's sufficiently close. 00:20:22.690 --> 00:20:24.308 align:middle line:84% And if they're in different clauses, 00:20:24.308 --> 00:20:25.600 align:middle line:90% they're not sufficiently close. 00:20:25.600 --> 00:20:26.980 align:middle line:90% And we'll talk about that soon. 00:20:26.980 --> 00:20:28.750 align:middle line:84% You'll get to look back with nostalgia 00:20:28.750 --> 00:20:32.920 align:middle line:84% on this little interchange in just a few minutes. 00:20:32.920 --> 00:20:33.420 align:middle line:90% Raquel? 00:20:33.420 --> 00:20:35.045 align:middle line:84% AUDIENCE: This isn't really a question, 00:20:35.045 --> 00:20:37.190 align:middle line:84% but I think it's interesting that coreference 00:20:37.190 --> 00:20:40.070 align:middle line:90% could create a presupposition. 00:20:40.070 --> 00:20:44.390 align:middle line:84% "The magician made herself disappear." 00:20:44.390 --> 00:20:46.040 align:middle line:84% It's like, OK, well, suddenly the fact 00:20:46.040 --> 00:20:49.985 align:middle line:84% that the magician is female is a presupposition that's there. 00:20:49.985 --> 00:20:53.110 align:middle line:84% But it could be wrong, and you couldn't star it necessarily 00:20:53.110 --> 00:20:54.110 align:middle line:90% because you're not sure. 00:20:54.110 --> 00:20:55.400 align:middle line:90% NORVIN RICHARDS: Yeah. 00:20:55.400 --> 00:20:56.420 align:middle line:90% No, I take your point. 00:20:56.420 --> 00:20:59.720 align:middle line:84% So I take your point to be, if you say "The magician made 00:20:59.720 --> 00:21:02.240 align:middle line:84% herself disappear," or "The magician made 00:21:02.240 --> 00:21:04.490 align:middle line:84% her father disappear," whether you 00:21:04.490 --> 00:21:06.813 align:middle line:84% use an anaphor or a pronoun, "magician" doesn't 00:21:06.813 --> 00:21:08.480 align:middle line:84% have any information about gender in it. 00:21:08.480 --> 00:21:11.420 align:middle line:84% But a fact about-- a thing about pronouns and reflexives 00:21:11.420 --> 00:21:14.420 align:middle line:84% in English is that they carry presuppositions 00:21:14.420 --> 00:21:15.920 align:middle line:84% about the gender of their reference. 00:21:15.920 --> 00:21:16.970 align:middle line:90% That's absolutely right. 00:21:16.970 --> 00:21:17.840 align:middle line:90% Yeah. 00:21:17.840 --> 00:21:19.400 align:middle line:90% Yeah, that's quite true. 00:21:19.400 --> 00:21:20.742 align:middle line:90% Yup. 00:21:20.742 --> 00:21:22.848 align:middle line:90% OK. 00:21:22.848 --> 00:21:23.890 align:middle line:90% All right, so-- oh, good. 00:21:23.890 --> 00:21:25.610 align:middle line:84% I did say it on the next slide, bind. 00:21:25.610 --> 00:21:29.333 align:middle line:84% This is just a name for c-command and "corefer with" 00:21:29.333 --> 00:21:31.750 align:middle line:84% just because we get tired of saying c-command and "corefer 00:21:31.750 --> 00:21:32.250 align:middle line:90% with." 00:21:32.250 --> 00:21:34.960 align:middle line:84% We say alpha binds beta if alpha c-commands 00:21:34.960 --> 00:21:36.010 align:middle line:90% and corefers with beta. 00:21:36.010 --> 00:21:40.810 align:middle line:84% So in "Susan must like herself," "Susan" binds "herself." 00:21:40.810 --> 00:21:45.580 align:middle line:84% "Susan" corefers with "herself," and it c-commands "herself." 00:21:45.580 --> 00:21:47.590 align:middle line:90% And anaphors need to be bound. 00:21:47.590 --> 00:21:50.585 align:middle line:84% And thanks to Faith, you know that, pretty soon, it's 00:21:50.585 --> 00:21:52.210 align:middle line:84% going to be more complicated than that. 00:21:52.210 --> 00:21:53.918 align:middle line:84% They need to be bound by something that's 00:21:53.918 --> 00:21:55.690 align:middle line:90% sufficiently close to them. 00:21:55.690 --> 00:21:57.870 align:middle line:90% Yeah. 00:21:57.870 --> 00:22:00.150 align:middle line:84% AUDIENCE: Speaking about behavior of c-command, 00:22:00.150 --> 00:22:04.116 align:middle line:84% you mentioned earlier in the semester how normal pronouns 00:22:04.116 --> 00:22:04.810 align:middle line:90% or-- 00:22:04.810 --> 00:22:07.383 align:middle line:84% yeah, normal pronouns are not A c-commanded. 00:22:07.383 --> 00:22:09.800 align:middle line:84% NORVIN RICHARDS: Yeah, so we're going to do pronouns next. 00:22:09.800 --> 00:22:10.100 align:middle line:90% Yeah. 00:22:10.100 --> 00:22:10.640 align:middle line:90% AUDIENCE: OK. 00:22:10.640 --> 00:22:11.630 align:middle line:84% NORVIN RICHARDS: Yeah, that's right. 00:22:11.630 --> 00:22:12.740 align:middle line:90% We're going to get to that. 00:22:12.740 --> 00:22:14.323 align:middle line:84% Actually, I think we're doing it next. 00:22:14.323 --> 00:22:15.620 align:middle line:90% Anaphors must be bound. 00:22:15.620 --> 00:22:17.700 align:middle line:90% OK, sorry, one more thing. 00:22:17.700 --> 00:22:19.110 align:middle line:90% There are two kinds of anaphors. 00:22:19.110 --> 00:22:21.027 align:middle line:84% There are reflexives, which are the ones we've 00:22:21.027 --> 00:22:23.302 align:middle line:84% been concentrating on, words like herself and himself. 00:22:23.302 --> 00:22:25.260 align:middle line:84% And then there are what are called reciprocals, 00:22:25.260 --> 00:22:26.468 align:middle line:90% which have the same behavior. 00:22:26.468 --> 00:22:28.080 align:middle line:90% They're words like "each other." 00:22:28.080 --> 00:22:30.410 align:middle line:84% So "John and Bill like each other." 00:22:30.410 --> 00:22:34.100 align:middle line:84% The meaning of "each other" is much more complicated 00:22:34.100 --> 00:22:36.020 align:middle line:84% than the meaning of "themselves," 00:22:36.020 --> 00:22:37.670 align:middle line:84% but "each other" and "themselves" 00:22:37.670 --> 00:22:40.700 align:middle line:84% have the same behavior in that they are OK in sentences 00:22:40.700 --> 00:22:43.130 align:middle line:84% like "John and Bill like each other" or "themselves," 00:22:43.130 --> 00:22:44.930 align:middle line:84% and that they are both bad in "John 00:22:44.930 --> 00:22:47.570 align:middle line:84% and Bill's father likes each other" or "themselves." 00:22:47.570 --> 00:22:50.870 align:middle line:84% So "each other" and "themselves" are both anaphors, 00:22:50.870 --> 00:22:55.203 align:middle line:84% and they're both subject to this requirement that they be bound. 00:22:55.203 --> 00:22:56.870 align:middle line:84% So "John and Bill like each other" means 00:22:56.870 --> 00:23:00.110 align:middle line:84% something like "John likes Bill and Bill likes John," yeah? 00:23:00.110 --> 00:23:02.750 align:middle line:84% There's literature on exactly what reciprocals 00:23:02.750 --> 00:23:05.730 align:middle line:90% mean because it's complicated. 00:23:05.730 --> 00:23:07.910 align:middle line:84% For example, it's possible to say things like, 00:23:07.910 --> 00:23:11.610 align:middle line:84% "We piled the books on top of each other." 00:23:11.610 --> 00:23:15.260 align:middle line:84% So each other doesn't mean for every pair 00:23:15.260 --> 00:23:19.803 align:middle line:84% in the group A relation B and B relation A, right? 00:23:19.803 --> 00:23:21.470 align:middle line:84% You might have thought, looking at this, 00:23:21.470 --> 00:23:23.012 align:middle line:84% that "John and Bill like each other," 00:23:23.012 --> 00:23:25.490 align:middle line:84% to interpret that, what you do is you take all of the pairs 00:23:25.490 --> 00:23:26.420 align:middle line:90% you can make out of the group. 00:23:26.420 --> 00:23:27.920 align:middle line:90% OK, there's one pair, you know? 00:23:27.920 --> 00:23:31.700 align:middle line:84% And what this means is John likes Bill and Bill likes John. 00:23:31.700 --> 00:23:34.620 align:middle line:84% But if you say "The books are piled on top of each other," 00:23:34.620 --> 00:23:38.930 align:middle line:84% it doesn't mean this book is on top of that book and-- 00:23:38.930 --> 00:23:40.215 align:middle line:90% That's not what it means. 00:23:40.215 --> 00:23:41.840 align:middle line:84% So there's work on trying to figure out 00:23:41.840 --> 00:23:44.150 align:middle line:84% what the heck it means, what reciprocals 00:23:44.150 --> 00:23:49.640 align:middle line:84% mean exactly, which we will not get any further into now. 00:23:49.640 --> 00:23:53.000 align:middle line:84% OK, this was Joseph's question, pronouns. 00:23:53.000 --> 00:23:54.480 align:middle line:90% Anaphors must be bound. 00:23:54.480 --> 00:23:55.633 align:middle line:90% Pronouns must be free. 00:23:55.633 --> 00:23:57.050 align:middle line:84% It sounds like a political slogan, 00:23:57.050 --> 00:23:59.480 align:middle line:84% but it's a fact about grammar, yeah? 00:23:59.480 --> 00:24:03.270 align:middle line:84% So pronouns need to not be bound. 00:24:03.270 --> 00:24:05.570 align:middle line:84% So "Susan likes herself" is fine. 00:24:05.570 --> 00:24:07.257 align:middle line:84% "Susan's father likes herself" is bad. 00:24:07.257 --> 00:24:08.840 align:middle line:84% That's the behavior of anaphors, which 00:24:08.840 --> 00:24:10.550 align:middle line:90% we're familiar with by now. 00:24:10.550 --> 00:24:13.190 align:middle line:84% And "her" behaves the same way only backwards. 00:24:13.190 --> 00:24:17.600 align:middle line:84% So "Susan likes her," fine sentence as long as "Susan" 00:24:17.600 --> 00:24:19.070 align:middle line:90% and "her" are different people. 00:24:19.070 --> 00:24:21.823 align:middle line:84% So "her" can't be bound by "Susan." 00:24:21.823 --> 00:24:23.240 align:middle line:84% It can't be c-commanded by "Susan" 00:24:23.240 --> 00:24:26.520 align:middle line:84% if "Susan" and "her" are going to refer to the same person. 00:24:26.520 --> 00:24:30.470 align:middle line:84% On the other hand, "Susan's father likes her"-- fine. 00:24:30.470 --> 00:24:34.770 align:middle line:84% So it's not quite true, but it's almost true 00:24:34.770 --> 00:24:37.410 align:middle line:84% that anaphors and pronouns are in complementary distribution 00:24:37.410 --> 00:24:38.330 align:middle line:90% with each other. 00:24:38.330 --> 00:24:41.340 align:middle line:84% So for a given meaning, you either use the pronoun, 00:24:41.340 --> 00:24:43.980 align:middle line:90% or you use the anaphor, OK? 00:24:43.980 --> 00:24:47.445 align:middle line:90% 00:24:47.445 --> 00:24:47.945 align:middle line:90% OK. 00:24:47.945 --> 00:24:50.840 align:middle line:90% 00:24:50.840 --> 00:24:54.790 align:middle line:84% More examples-- even more examples. 00:24:54.790 --> 00:24:58.560 align:middle line:84% So now, we're getting to Faith's point-- 00:24:58.560 --> 00:25:01.050 align:middle line:84% my version of Faith's point, sorry. 00:25:01.050 --> 00:25:03.960 align:middle line:90% So "Susan likes herself." 00:25:03.960 --> 00:25:05.310 align:middle line:90% We've seen this is OK. 00:25:05.310 --> 00:25:06.870 align:middle line:90% "Susan thinks I like herself." 00:25:06.870 --> 00:25:09.840 align:middle line:84% Or I have the sentence on the board, 00:25:09.840 --> 00:25:12.300 align:middle line:84% "Mary thinks Susan must like herself." 00:25:12.300 --> 00:25:16.740 align:middle line:84% "Herself" is trying to get bound, in this case, 00:25:16.740 --> 00:25:18.390 align:middle line:90% on the slide by Susan. 00:25:18.390 --> 00:25:20.670 align:middle line:90% And it can't be. 00:25:20.670 --> 00:25:22.260 align:middle line:84% The kinds of trees we were drawing 00:25:22.260 --> 00:25:24.150 align:middle line:84% for this kind of sentence, Susan certainly 00:25:24.150 --> 00:25:25.740 align:middle line:90% does c-command herself. 00:25:25.740 --> 00:25:29.660 align:middle line:84% I don't think I put the tree on the slide. 00:25:29.660 --> 00:25:31.890 align:middle line:84% So let me just create it quickly. 00:25:31.890 --> 00:25:42.670 align:middle line:84% "Susan thinks"-- there's going to be a CP here, then null C. 00:25:42.670 --> 00:25:45.430 align:middle line:84% And then down here, we have a noun phrase, 00:25:45.430 --> 00:25:52.990 align:middle line:84% "I," and a verb phrase like "herself." 00:25:52.990 --> 00:25:56.450 align:middle line:90% So if we ask ourselves does-- 00:25:56.450 --> 00:25:57.640 align:middle line:90% here's the tree. 00:25:57.640 --> 00:25:59.140 align:middle line:90% Did I write the right sentence? 00:25:59.140 --> 00:26:01.720 align:middle line:84% "Susan thinks I like herself," yes. 00:26:01.720 --> 00:26:04.930 align:middle line:84% If we ask ourself, does "Susan" c-command "herself," 00:26:04.930 --> 00:26:06.520 align:middle line:90% the answer is yes. 00:26:06.520 --> 00:26:08.710 align:middle line:84% If we ask what are the nodes that 00:26:08.710 --> 00:26:11.602 align:middle line:84% dominate the noun phrase "Susan," well, 00:26:11.602 --> 00:26:12.310 align:middle line:90% there's only one. 00:26:12.310 --> 00:26:14.200 align:middle line:90% It's this TP. 00:26:14.200 --> 00:26:16.300 align:middle line:84% And so if we're asking, does "Susan" c-command 00:26:16.300 --> 00:26:18.627 align:middle line:84% "herself," well, we're really asking, 00:26:18.627 --> 00:26:19.960 align:middle line:90% does this TP dominate "herself"? 00:26:19.960 --> 00:26:21.730 align:middle line:90% And yes, it does. 00:26:21.730 --> 00:26:25.800 align:middle line:90% So "Susan" c-commands "herself." 00:26:25.800 --> 00:26:29.100 align:middle line:84% So it looks as though "Susan" ought to bind "herself." 00:26:29.100 --> 00:26:31.800 align:middle line:84% What we're learning here is the complication 00:26:31.800 --> 00:26:35.865 align:middle line:84% that Faith made me reveal early, which is that not only do 00:26:35.865 --> 00:26:36.990 align:middle line:90% anaphors need to get bound. 00:26:36.990 --> 00:26:38.070 align:middle line:84% They need to get bound by something 00:26:38.070 --> 00:26:38.987 align:middle line:90% which is close enough. 00:26:38.987 --> 00:26:42.000 align:middle line:90% 00:26:42.000 --> 00:26:43.560 align:middle line:90% There's a principle. 00:26:43.560 --> 00:26:45.420 align:middle line:84% It's called principle A. You can think 00:26:45.420 --> 00:26:47.190 align:middle line:90% of A as standing for Anaphor. 00:26:47.190 --> 00:26:52.320 align:middle line:84% It says anaphors must be bound within TP. 00:26:52.320 --> 00:26:57.270 align:middle line:84% If this were Intro to Syntax, we would now spend weeks, possibly 00:26:57.270 --> 00:27:01.350 align:middle line:84% months trying to figure out what the real locality 00:27:01.350 --> 00:27:03.720 align:middle line:84% requirement is because it's more complicated than that. 00:27:03.720 --> 00:27:05.370 align:middle line:90% But that's close. 00:27:05.370 --> 00:27:10.680 align:middle line:84% So the idea here is that "herself," yes, it's 00:27:10.680 --> 00:27:12.720 align:middle line:90% bound by "Susan" in this tree. 00:27:12.720 --> 00:27:15.690 align:middle line:84% But it's not bound by "Susan" within the smallest TP. 00:27:15.690 --> 00:27:18.150 align:middle line:84% Here's the smallest TP that includes "herself," 00:27:18.150 --> 00:27:21.800 align:middle line:90% and "Susan" is too far away. 00:27:21.800 --> 00:27:25.715 align:middle line:84% So there's a locality requirement. 00:27:25.715 --> 00:27:26.590 align:middle line:90% Does that make sense? 00:27:26.590 --> 00:27:28.510 align:middle line:84% So principle A says, anaphors need 00:27:28.510 --> 00:27:31.110 align:middle line:84% to be bound within the smallest TP that contains them. 00:27:31.110 --> 00:27:33.920 align:middle line:90% 00:27:33.920 --> 00:27:37.430 align:middle line:84% And then principle B, where B stands for the principle that 00:27:37.430 --> 00:27:40.850 align:middle line:84% comes after principle A, says, pronouns 00:27:40.850 --> 00:27:44.870 align:middle line:84% must be free within TP, so just as we saw before, 00:27:44.870 --> 00:27:46.700 align:middle line:84% that pronouns and anaphors are, so 00:27:46.700 --> 00:27:49.520 align:middle line:84% far, in complementary distribution for a given 00:27:49.520 --> 00:27:50.030 align:middle line:90% reading. 00:27:50.030 --> 00:27:52.340 align:middle line:84% So you can say "Susan likes herself," 00:27:52.340 --> 00:27:55.340 align:middle line:84% and that means that you can't say "Susan likes her" 00:27:55.340 --> 00:27:56.930 align:middle line:84% and mean the same thing, that you mean 00:27:56.930 --> 00:27:59.930 align:middle line:90% that "Susan likes herself." 00:27:59.930 --> 00:28:03.650 align:middle line:84% Similarly, we've now seen, thanks to principle A, 00:28:03.650 --> 00:28:07.200 align:middle line:84% that you cannot say "Susan thinks I like herself," 00:28:07.200 --> 00:28:11.290 align:middle line:84% where "herself" refers back to "Susan." 00:28:11.290 --> 00:28:14.110 align:middle line:84% And principle B says, yeah, pronouns 00:28:14.110 --> 00:28:15.880 align:middle line:90% have to be free within TP. 00:28:15.880 --> 00:28:18.430 align:middle line:84% So "Susan thinks I like her," where "her" refers to "Susan." 00:28:18.430 --> 00:28:20.640 align:middle line:84% Fine, it doesn't have to refer to "Susan," right? 00:28:20.640 --> 00:28:21.640 align:middle line:90% It can refer to anybody. 00:28:21.640 --> 00:28:25.090 align:middle line:84% It can be free regardless of whether it's bound by "Susan" 00:28:25.090 --> 00:28:26.770 align:middle line:90% or not, free within TP. 00:28:26.770 --> 00:28:27.460 align:middle line:90% Yeah, Faith. 00:28:27.460 --> 00:28:28.960 align:middle line:84% AUDIENCE: I may have just forgotten, 00:28:28.960 --> 00:28:33.250 align:middle line:84% but with that T there, is the invisible T that's dominated 00:28:33.250 --> 00:28:35.080 align:middle line:90% by T bar always necessary? 00:28:35.080 --> 00:28:36.400 align:middle line:90% Or could you just say-- 00:28:36.400 --> 00:28:36.910 align:middle line:90% NORVIN RICHARDS: This? 00:28:36.910 --> 00:28:37.535 align:middle line:90% AUDIENCE: Yeah. 00:28:37.535 --> 00:28:38.040 align:middle line:90% And-- 00:28:38.040 --> 00:28:39.370 align:middle line:90% NORVIN RICHARDS: And also this? 00:28:39.370 --> 00:28:39.870 align:middle line:90% Yeah. 00:28:39.870 --> 00:28:40.460 align:middle line:90% You know what? 00:28:40.460 --> 00:28:42.252 align:middle line:84% In most of the trees I've been showing you, 00:28:42.252 --> 00:28:45.305 align:middle line:84% I've been smart enough to put auxiliaries in the sentences. 00:28:45.305 --> 00:28:46.930 align:middle line:84% So I've been saying things like, "Susan 00:28:46.930 --> 00:28:50.650 align:middle line:84% will think that I must like herself." 00:28:50.650 --> 00:28:53.980 align:middle line:84% And then people don't ask the question that you're asking. 00:28:53.980 --> 00:28:57.400 align:middle line:84% But when I drew this tree, I was not smart enough to do that. 00:28:57.400 --> 00:29:01.933 align:middle line:84% There does need to be a T because-- 00:29:01.933 --> 00:29:03.100 align:middle line:90% why does there need to be T? 00:29:03.100 --> 00:29:06.850 align:middle line:84% Well, we have principles like the extended projection 00:29:06.850 --> 00:29:10.330 align:middle line:84% principle, which says that TP must have a specifier. 00:29:10.330 --> 00:29:13.960 align:middle line:84% And sentences without auxiliaries in them 00:29:13.960 --> 00:29:15.905 align:middle line:90% show the effects of that. 00:29:15.905 --> 00:29:18.280 align:middle line:84% So it doesn't matter whether there's an auxiliary or not. 00:29:18.280 --> 00:29:22.452 align:middle line:90% 00:29:22.452 --> 00:29:22.952 align:middle line:90% Yeah. 00:29:22.952 --> 00:29:24.770 align:middle line:84% It doesn't matter whether there's an auxiliary or not. 00:29:24.770 --> 00:29:27.395 align:middle line:84% You still get the effects of the extended projection principle. 00:29:27.395 --> 00:29:29.650 align:middle line:84% So we do want there to be a T because we 00:29:29.650 --> 00:29:30.598 align:middle line:90% want there to be a TP. 00:29:30.598 --> 00:29:32.890 align:middle line:84% And we don't have any way to make a TP without starting 00:29:32.890 --> 00:29:36.670 align:middle line:84% with a T. So you have to merge T with a verb phrase and so on. 00:29:36.670 --> 00:29:39.376 align:middle line:90% AUDIENCE: [INAUDIBLE] 00:29:39.376 --> 00:29:40.533 align:middle line:90% 00:29:40.533 --> 00:29:41.450 align:middle line:90% NORVIN RICHARDS: Yeah. 00:29:41.450 --> 00:29:42.500 align:middle line:90% So I mean, here's-- 00:29:42.500 --> 00:29:45.110 align:middle line:90% 00:29:45.110 --> 00:29:47.170 align:middle line:84% I'm not totally unhappy about the direction 00:29:47.170 --> 00:29:48.920 align:middle line:84% this conversation is taking because you're 00:29:48.920 --> 00:29:52.940 align:middle line:84% forcing me to talk about syntax, which is fine, even though this 00:29:52.940 --> 00:29:54.660 align:middle line:84% is the semantic part of the story. 00:29:54.660 --> 00:29:58.380 align:middle line:84% Suppose I was smart enough to use an auxiliary here. 00:29:58.380 --> 00:30:00.530 align:middle line:90% So we'll do "Susan will think." 00:30:00.530 --> 00:30:03.535 align:middle line:84% And we'll forget about the embedded clause for a second. 00:30:03.535 --> 00:30:04.910 align:middle line:84% We've talked about this, I think, 00:30:04.910 --> 00:30:09.620 align:middle line:84% that, if you want to ask a yes/no question, 00:30:09.620 --> 00:30:13.370 align:middle line:84% the way to do it is to head move T into C. 00:30:13.370 --> 00:30:19.430 align:middle line:84% So you ask questions like, "Will Susan think that" blah-di-blah? 00:30:19.430 --> 00:30:21.630 align:middle line:90% So bear that in mind. 00:30:21.630 --> 00:30:25.220 align:middle line:84% And now, let's switch to the version of this 00:30:25.220 --> 00:30:26.930 align:middle line:90% where there isn't an auxiliary. 00:30:26.930 --> 00:30:29.970 align:middle line:90% "Susan thinks that I like." 00:30:29.970 --> 00:30:32.570 align:middle line:84% If I want to make a question out of that, what I do 00:30:32.570 --> 00:30:35.000 align:middle line:90% is I put "Does" here. 00:30:35.000 --> 00:30:36.630 align:middle line:90% And I get rid of the S here. 00:30:36.630 --> 00:30:40.400 align:middle line:84% So it's "Does Susan think that" blah-di-blah? 00:30:40.400 --> 00:30:42.540 align:middle line:84% So there's a story about what's going on here, 00:30:42.540 --> 00:30:45.740 align:middle line:84% which says what's in T-- so there are various things that 00:30:45.740 --> 00:30:50.840 align:middle line:84% can be in T. There's "will" and "should," "might." 00:30:50.840 --> 00:30:56.750 align:middle line:84% And there's also "-s," as in "Mary thinks that I like her." 00:30:56.750 --> 00:31:01.382 align:middle line:84% And if you do T to C of "will" or "should" or "might," 00:31:01.382 --> 00:31:02.840 align:middle line:84% then you just move them over there. 00:31:02.840 --> 00:31:04.040 align:middle line:90% And nothing else happens. 00:31:04.040 --> 00:31:06.960 align:middle line:84% If you have a "-s" then there are a couple of things that 00:31:06.960 --> 00:31:07.460 align:middle line:90% happen. 00:31:07.460 --> 00:31:10.430 align:middle line:84% If you aren't head-moving it, then it attaches to this verb. 00:31:10.430 --> 00:31:11.920 align:middle line:84% So you end up with "Susan thinks." 00:31:11.920 --> 00:31:14.420 align:middle line:84% And we have to figure out how these two things come together 00:31:14.420 --> 00:31:15.380 align:middle line:90% to form a verb. 00:31:15.380 --> 00:31:20.360 align:middle line:84% And sadly, it's long enough since we did syntax 00:31:20.360 --> 00:31:23.000 align:middle line:84% that I'm unwilling to try to tell you that. 00:31:23.000 --> 00:31:25.310 align:middle line:84% There's a big literature on trying to figure that out. 00:31:25.310 --> 00:31:27.902 align:middle line:84% And then the other story is, if you're going to do T to C, 00:31:27.902 --> 00:31:29.360 align:middle line:84% if you're going to make a question, 00:31:29.360 --> 00:31:30.470 align:middle line:90% you're going to take this. 00:31:30.470 --> 00:31:32.090 align:middle line:90% It's not going to show up here. 00:31:32.090 --> 00:31:33.960 align:middle line:90% It's going to show up up here. 00:31:33.960 --> 00:31:37.400 align:middle line:84% So you would have expected this to be "-s Susan think I like 00:31:37.400 --> 00:31:38.060 align:middle line:90% herself?" 00:31:38.060 --> 00:31:41.050 align:middle line:84% But "-s" is too small to be an English word or something like 00:31:41.050 --> 00:31:41.550 align:middle line:90% that. 00:31:41.550 --> 00:31:45.030 align:middle line:90% And so you put in this "do." 00:31:45.030 --> 00:31:48.110 align:middle line:84% There's a process that is called "do support." 00:31:48.110 --> 00:31:51.480 align:middle line:84% You add "do" there so that we'll have a verb to be part of. 00:31:51.480 --> 00:31:53.300 align:middle line:90% It's the story that people tell. 00:31:53.300 --> 00:31:55.550 align:middle line:84% And so that's why the question version of that 00:31:55.550 --> 00:31:58.100 align:middle line:84% is "Does Susan think that I like her?" 00:31:58.100 --> 00:32:01.220 align:middle line:84% Can you see why I have been hiding all 00:32:01.220 --> 00:32:03.200 align:middle line:90% of this from you all semester? 00:32:03.200 --> 00:32:05.660 align:middle line:84% So I've been using auxiliaries instead of giving you verbs 00:32:05.660 --> 00:32:07.100 align:middle line:90% that have suffixes on them. 00:32:07.100 --> 00:32:09.898 align:middle line:84% This is why, because I didn't want to do this. 00:32:09.898 --> 00:32:10.940 align:middle line:90% But it's been fun anyway. 00:32:10.940 --> 00:32:12.440 align:middle line:84% But now, it's over, except it's not. 00:32:12.440 --> 00:32:14.325 align:middle line:90% Joseph? 00:32:14.325 --> 00:32:15.700 align:middle line:84% AUDIENCE: Just a brief question-- 00:32:15.700 --> 00:32:15.827 align:middle line:90% NORVIN RICHARDS: Yes. 00:32:15.827 --> 00:32:17.230 align:middle line:84% AUDIENCE: --about that movement, you 00:32:17.230 --> 00:32:19.240 align:middle line:84% said the extended production principle requires 00:32:19.240 --> 00:32:21.040 align:middle line:90% you TP has a specifier. 00:32:21.040 --> 00:32:24.960 align:middle line:84% So that means that, because [INAUDIBLE] creating 00:32:24.960 --> 00:32:28.430 align:middle line:84% a yes or no question means moving ahead out of TP, 00:32:28.430 --> 00:32:30.400 align:middle line:90% you can't move "Susan." 00:32:30.400 --> 00:32:32.453 align:middle line:84% Because then TP was the specifier, right? 00:32:32.453 --> 00:32:33.370 align:middle line:90% NORVIN RICHARDS: Yeah. 00:32:33.370 --> 00:32:34.180 align:middle line:90% Oh, I see. 00:32:34.180 --> 00:32:34.900 align:middle line:90% AUDIENCE: Yeah, why [INAUDIBLE]? 00:32:34.900 --> 00:32:35.800 align:middle line:90% NORVIN RICHARDS: No, no. 00:32:35.800 --> 00:32:37.508 align:middle line:84% Oh, you mean why are you moving the head? 00:32:37.508 --> 00:32:39.580 align:middle line:90% Why are you not moving a phrase? 00:32:39.580 --> 00:32:41.747 align:middle line:84% AUDIENCE: Why can't you just move the phrase around? 00:32:41.747 --> 00:32:43.540 align:middle line:90% 00:32:43.540 --> 00:32:46.120 align:middle line:84% NORVIN RICHARDS: There isn't a deep answer to that question. 00:32:46.120 --> 00:32:48.730 align:middle line:84% Forming questions involves taking whatever is in T 00:32:48.730 --> 00:32:51.550 align:middle line:84% and moving it into C. You'd like to know 00:32:51.550 --> 00:32:56.100 align:middle line:84% why we don't move other things, and that's a good question. 00:32:56.100 --> 00:32:56.730 align:middle line:90% This is-- 00:32:56.730 --> 00:32:57.696 align:middle line:90% AUDIENCE: [INAUDIBLE] to her. 00:32:57.696 --> 00:33:00.070 align:middle line:84% And I said, you can't do this because of the shortest path. 00:33:00.070 --> 00:33:00.580 align:middle line:90% NORVIN RICHARDS: Ah, yes. 00:33:00.580 --> 00:33:00.880 align:middle line:90% I see. 00:33:00.880 --> 00:33:01.180 align:middle line:90% Yes. 00:33:01.180 --> 00:33:03.650 align:middle line:84% So, yeah, you can't take the verb and move it up there. 00:33:03.650 --> 00:33:05.192 align:middle line:84% That's right because that's too long. 00:33:05.192 --> 00:33:07.540 align:middle line:90% Why can't you move this? 00:33:07.540 --> 00:33:09.040 align:middle line:90% Because it's not a head. 00:33:09.040 --> 00:33:10.960 align:middle line:90% It's a phrase. 00:33:10.960 --> 00:33:13.060 align:middle line:84% Apparently, we think that only heads can 00:33:13.060 --> 00:33:14.480 align:middle line:90% move to attach to other heads. 00:33:14.480 --> 00:33:14.980 align:middle line:90% Yeah. 00:33:14.980 --> 00:33:16.692 align:middle line:90% And we might want to know why. 00:33:16.692 --> 00:33:18.650 align:middle line:84% There are lots of things we might want to know. 00:33:18.650 --> 00:33:20.080 align:middle line:90% AUDIENCE: [INAUDIBLE] 00:33:20.080 --> 00:33:20.650 align:middle line:90% 00:33:20.650 --> 00:33:22.270 align:middle line:90% NORVIN RICHARDS: Go take syntax. 00:33:22.270 --> 00:33:24.310 align:middle line:90% Yeah. 00:33:24.310 --> 00:33:25.780 align:middle line:90% Yeah? 00:33:25.780 --> 00:33:26.830 align:middle line:90% OK. 00:33:26.830 --> 00:33:29.650 align:middle line:84% All right, all of you, you dragged 00:33:29.650 --> 00:33:31.990 align:middle line:84% me kicking and screaming into doing more syntax. 00:33:31.990 --> 00:33:32.560 align:middle line:90% Shame on you. 00:33:32.560 --> 00:33:35.350 align:middle line:90% OK. 00:33:35.350 --> 00:33:37.090 align:middle line:84% Right, principle A, principle B, yes? 00:33:37.090 --> 00:33:39.940 align:middle line:84% Are people reasonably happy with principle A and principle B? 00:33:39.940 --> 00:33:42.220 align:middle line:84% So they say anaphors need to be bound within TP. 00:33:42.220 --> 00:33:43.690 align:middle line:84% Pronouns have to be free within TP. 00:33:43.690 --> 00:33:46.800 align:middle line:90% 00:33:46.800 --> 00:33:50.520 align:middle line:84% Side note, since I just told you universal about anaphors, 00:33:50.520 --> 00:33:52.290 align:middle line:84% anaphors only ever care about c-command. 00:33:52.290 --> 00:33:54.715 align:middle line:84% They never seem to care about linear order. 00:33:54.715 --> 00:33:57.090 align:middle line:84% They don't care about what precedes them or follows them. 00:33:57.090 --> 00:33:59.220 align:middle line:84% I should say that there are points 00:33:59.220 --> 00:34:00.780 align:middle line:90% of variation between languages. 00:34:00.780 --> 00:34:04.620 align:middle line:84% There are languages that have anaphors that need to be bound, 00:34:04.620 --> 00:34:07.620 align:middle line:84% but they can be bound longer than this. 00:34:07.620 --> 00:34:09.330 align:middle line:84% So there are languages that have anaphors 00:34:09.330 --> 00:34:10.980 align:middle line:84% that can be bound outside of the TP 00:34:10.980 --> 00:34:13.270 align:middle line:84% that they're in called long-distance anaphors. 00:34:13.270 --> 00:34:16.739 align:middle line:84% English doesn't have them, but Mandarin does, for example. 00:34:16.739 --> 00:34:18.989 align:middle line:84% So Mandarin, [NON-ENGLISH] is a long-distance anaphor. 00:34:18.989 --> 00:34:22.739 align:middle line:90% 00:34:22.739 --> 00:34:25.500 align:middle line:84% Principle A is not a parameter that distinguishes languages. 00:34:25.500 --> 00:34:27.420 align:middle line:84% It's a parameter that distinguishes anaphors. 00:34:27.420 --> 00:34:29.722 align:middle line:84% So Mandarin actually has two anaphors. 00:34:29.722 --> 00:34:31.430 align:middle line:84% It has an anaphor that behaves like this. 00:34:31.430 --> 00:34:33.600 align:middle line:84% It's just pronounced [NON-ENGLISH].. 00:34:33.600 --> 00:34:37.170 align:middle line:84% And then it has another one that is long distance that can 00:34:37.170 --> 00:34:38.610 align:middle line:90% be bound from outside its TP. 00:34:38.610 --> 00:34:41.040 align:middle line:84% It's just [NON-ENGLISH] by itself. 00:34:41.040 --> 00:34:44.310 align:middle line:84% Those of you who speak Mandarin can be thinking about that. 00:34:44.310 --> 00:34:47.040 align:middle line:84% There are languages that have more anaphors than that. 00:34:47.040 --> 00:34:48.659 align:middle line:84% Korean, I think, holds the record. 00:34:48.659 --> 00:34:50.909 align:middle line:84% It has four different anaphors with slightly different 00:34:50.909 --> 00:34:51.690 align:middle line:90% behaviors. 00:34:51.690 --> 00:34:52.190 align:middle line:90% Yeah. 00:34:52.190 --> 00:34:54.600 align:middle line:84% AUDIENCE: Do languages that have long-distance anaphors 00:34:54.600 --> 00:34:59.490 align:middle line:84% also always have this type of anaphor, or does it vary? 00:34:59.490 --> 00:35:03.120 align:middle line:84% NORVIN RICHARDS: As far as I know, 00:35:03.120 --> 00:35:05.970 align:middle line:84% every language has this type of anaphor. 00:35:05.970 --> 00:35:12.370 align:middle line:84% And long-distance anaphors are an additional luxury 00:35:12.370 --> 00:35:14.222 align:middle line:84% that a language can allow itself or not. 00:35:14.222 --> 00:35:16.180 align:middle line:84% I don't think there are any languages that only 00:35:16.180 --> 00:35:17.830 align:middle line:90% have long-distance anaphors. 00:35:17.830 --> 00:35:20.570 align:middle line:90% Yeah, good question. 00:35:20.570 --> 00:35:21.070 align:middle line:90% Yup? 00:35:21.070 --> 00:35:21.570 align:middle line:90% OK. 00:35:21.570 --> 00:35:27.670 align:middle line:90% 00:35:27.670 --> 00:35:28.270 align:middle line:90% Where are we? 00:35:28.270 --> 00:35:29.020 align:middle line:90% OK, yes. 00:35:29.020 --> 00:35:32.980 align:middle line:84% So we have principle A. We have principle B. Anaphors need 00:35:32.980 --> 00:35:33.940 align:middle line:90% to be bound within TP. 00:35:33.940 --> 00:35:35.500 align:middle line:84% Pronouns have to be free within TP. 00:35:35.500 --> 00:35:38.020 align:middle line:84% And let's stick to English again. 00:35:38.020 --> 00:35:40.390 align:middle line:84% Now, let's consider these sentences, "She likes Susan" 00:35:40.390 --> 00:35:47.612 align:middle line:84% and "Her father likes Susan," with these subscripts 00:35:47.612 --> 00:35:49.320 align:middle line:84% and these grammaticality judgments, which 00:35:49.320 --> 00:35:50.530 align:middle line:90% I hope everybody agrees with. 00:35:50.530 --> 00:35:52.018 align:middle line:90% So "She likes Susan"-- 00:35:52.018 --> 00:35:54.435 align:middle line:84% fine sentence, but she and Susan can't be the same person. 00:35:54.435 --> 00:35:56.803 align:middle line:90% 00:35:56.803 --> 00:35:58.845 align:middle line:84% Does either of our principles handle these facts? 00:35:58.845 --> 00:36:03.380 align:middle line:90% 00:36:03.380 --> 00:36:07.910 align:middle line:84% That's a leading question if there ever was one. 00:36:07.910 --> 00:36:09.750 align:middle line:90% No, they do not. 00:36:09.750 --> 00:36:11.938 align:middle line:84% So first of all, principle A is obviously 00:36:11.938 --> 00:36:12.980 align:middle line:90% going to be useless here. 00:36:12.980 --> 00:36:15.560 align:middle line:84% There are no anaphors in these sentences. 00:36:15.560 --> 00:36:17.630 align:middle line:84% So forget principle A. Principle B, 00:36:17.630 --> 00:36:20.880 align:middle line:84% pronouns need to be free within TP. 00:36:20.880 --> 00:36:24.300 align:middle line:84% Well, is she free within TP in "She likes Susan"? 00:36:24.300 --> 00:36:30.040 align:middle line:84% Is there a noun phrase that c-commands "she" and corefers 00:36:30.040 --> 00:36:30.540 align:middle line:90% with it? 00:36:30.540 --> 00:36:34.990 align:middle line:90% 00:36:34.990 --> 00:36:37.220 align:middle line:90% Is "she" free? 00:36:37.220 --> 00:36:38.450 align:middle line:90% Yes. 00:36:38.450 --> 00:36:39.260 align:middle line:90% She is free. 00:36:39.260 --> 00:36:40.310 align:middle line:90% Shall I draw a tree? 00:36:40.310 --> 00:36:43.400 align:middle line:90% 00:36:43.400 --> 00:36:44.570 align:middle line:90% What's the sentence? 00:36:44.570 --> 00:36:45.320 align:middle line:90% "She likes Susan." 00:36:45.320 --> 00:36:50.350 align:middle line:90% 00:36:50.350 --> 00:36:54.700 align:middle line:84% Someday, I'm going to ask somebody 00:36:54.700 --> 00:36:58.090 align:middle line:84% to make me a stamp, a chalk stamp, you know, 00:36:58.090 --> 00:37:01.040 align:middle line:84% that just has this tree, TP, T bar, [INAUDIBLE].. 00:37:01.040 --> 00:37:01.540 align:middle line:90% Wham. 00:37:01.540 --> 00:37:03.000 align:middle line:90% And then there it would be. 00:37:03.000 --> 00:37:03.860 align:middle line:90% That would be great. 00:37:03.860 --> 00:37:05.950 align:middle line:90% "She likes Susan," yeah? 00:37:05.950 --> 00:37:06.910 align:middle line:90% Here's the pronoun. 00:37:06.910 --> 00:37:08.560 align:middle line:84% Pronoun needs to be free within TP. 00:37:08.560 --> 00:37:11.410 align:middle line:84% That is there had better not be anything c-commanding "she" 00:37:11.410 --> 00:37:12.580 align:middle line:90% that corefers with "she." 00:37:12.580 --> 00:37:13.990 align:middle line:90% Does anything c-command "she"? 00:37:13.990 --> 00:37:19.420 align:middle line:90% 00:37:19.420 --> 00:37:23.110 align:middle line:90% I guess maybe T bar. 00:37:23.110 --> 00:37:27.670 align:middle line:84% So T bar is dominated by all the same nodes as noun phrase. 00:37:27.670 --> 00:37:29.263 align:middle line:84% But T bar doesn't refer to "she." 00:37:29.263 --> 00:37:30.430 align:middle line:90% It doesn't refer to "Susan." 00:37:30.430 --> 00:37:31.870 align:middle line:90% So "she" is free, yeah. 00:37:31.870 --> 00:37:35.700 align:middle line:84% So our principles don't handle these facts. 00:37:35.700 --> 00:37:38.310 align:middle line:84% There are at least two things we could do. 00:37:38.310 --> 00:37:40.740 align:middle line:84% One would be to say, ah, principle B, 00:37:40.740 --> 00:37:46.650 align:middle line:84% we were wrong to confine principle B to pronouns. 00:37:46.650 --> 00:37:50.790 align:middle line:84% So it's really just a distinction between anaphors 00:37:50.790 --> 00:37:52.750 align:middle line:90% and everything else. 00:37:52.750 --> 00:37:56.070 align:middle line:84% So the problem with "She likes Susan" isn't "she." 00:37:56.070 --> 00:37:57.750 align:middle line:90% It's "Susan." 00:37:57.750 --> 00:38:00.870 align:middle line:84% So "Susan" also needs to be free within TP. 00:38:00.870 --> 00:38:03.150 align:middle line:84% And then the story would be "Her father likes Susan." 00:38:03.150 --> 00:38:05.530 align:middle line:84% "Her" is embedded inside "her father." 00:38:05.530 --> 00:38:07.840 align:middle line:84% And so it doesn't c-command "Susan," and so there. 00:38:07.840 --> 00:38:12.170 align:middle line:84% That version of principle B would cover this. 00:38:12.170 --> 00:38:21.920 align:middle line:84% But pronouns really are the opposite of anaphors. 00:38:21.920 --> 00:38:23.690 align:middle line:90% They need to be free within TP. 00:38:23.690 --> 00:38:27.650 align:middle line:84% So "Susan thinks I like her," where "Susan" c-commands "her" 00:38:27.650 --> 00:38:30.638 align:middle line:84% and corefers with "her," so "Susan" binds "her," is fine. 00:38:30.638 --> 00:38:32.180 align:middle line:84% Because although "her" is bound, it's 00:38:32.180 --> 00:38:34.290 align:middle line:84% bound by something that's far away. 00:38:34.290 --> 00:38:36.180 align:middle line:90% And that's all right. 00:38:36.180 --> 00:38:38.800 align:middle line:84% But in "She thinks I like Susan," 00:38:38.800 --> 00:38:42.480 align:middle line:84% we're seeing that names do not behave like pronouns. 00:38:42.480 --> 00:38:47.940 align:middle line:84% "Susan" requires absolute freedom, freedom everywhere. 00:38:47.940 --> 00:38:51.060 align:middle line:84% Pronouns just need to be free within a certain domain, 00:38:51.060 --> 00:38:52.500 align:middle line:90% within TP let's say. 00:38:52.500 --> 00:38:56.640 align:middle line:84% But a name like "Susan" needs to be free everywhere. 00:38:56.640 --> 00:38:58.650 align:middle line:84% The fact that "Susan" is bound by "she" 00:38:58.650 --> 00:39:00.900 align:middle line:84% in that second example, "She thinks I like Susan," 00:39:00.900 --> 00:39:02.207 align:middle line:90% means that the sentence is bad. 00:39:02.207 --> 00:39:03.915 align:middle line:84% It doesn't matter that they're far apart. 00:39:03.915 --> 00:39:07.990 align:middle line:90% 00:39:07.990 --> 00:39:11.440 align:middle line:84% So we make up a new principle-- it's called principle C-- 00:39:11.440 --> 00:39:14.410 align:middle line:84% and, also, a new term, R-expressions. 00:39:14.410 --> 00:39:19.720 align:middle line:84% R-expressions means things that are not anaphors or pronouns. 00:39:19.720 --> 00:39:22.940 align:middle line:84% It stands for Referring expressions, 00:39:22.940 --> 00:39:26.350 align:middle line:84% which is an unfortunate name because they don't necessarily 00:39:26.350 --> 00:39:27.248 align:middle line:90% refer. 00:39:27.248 --> 00:39:27.790 align:middle line:90% "Susan" does. 00:39:27.790 --> 00:39:28.582 align:middle line:90% It refers to Susan. 00:39:28.582 --> 00:39:31.600 align:middle line:90% But anyway, that's all it means. 00:39:31.600 --> 00:39:33.340 align:middle line:84% Things that are not anaphors or pronouns. 00:39:33.340 --> 00:39:36.910 align:middle line:84% So everything that isn't subject to principle A or principle B 00:39:36.910 --> 00:39:39.670 align:middle line:84% is subject to principle C. Every noun phrase that 00:39:39.670 --> 00:39:41.650 align:middle line:84% isn't subject to principle A or principal B 00:39:41.650 --> 00:39:44.590 align:middle line:84% is subject to principle C, which says our expressions must 00:39:44.590 --> 00:39:47.223 align:middle line:90% be free everywhere. 00:39:47.223 --> 00:39:49.390 align:middle line:84% Our expressions must be free, and everywhere they're 00:39:49.390 --> 00:39:50.540 align:middle line:90% in chains. 00:39:50.540 --> 00:39:53.650 align:middle line:90% Yeah, OK? 00:39:53.650 --> 00:39:54.430 align:middle line:90% All right. 00:39:54.430 --> 00:39:58.390 align:middle line:84% And then this is the fact that I made a big deal about before. 00:39:58.390 --> 00:40:02.860 align:middle line:84% Even though which thing precedes which other thing is 00:40:02.860 --> 00:40:05.470 align:middle line:84% the kind of thing that every infant knows as soon as they 00:40:05.470 --> 00:40:08.410 align:middle line:84% can hear, words and segment words, which infants 00:40:08.410 --> 00:40:13.300 align:middle line:84% are eerily good at, even though-- 00:40:13.300 --> 00:40:19.090 align:middle line:84% if you were going to design a computer, a language 00:40:19.090 --> 00:40:21.880 align:middle line:90% for computers to use-- 00:40:21.880 --> 00:40:23.380 align:middle line:90% I am not a computer person. 00:40:23.380 --> 00:40:25.030 align:middle line:90% You may have picked up on this. 00:40:25.030 --> 00:40:28.240 align:middle line:84% Even I think it would be easier for me to get a computer 00:40:28.240 --> 00:40:31.960 align:middle line:84% to figure out which thing linearly precedes which 00:40:31.960 --> 00:40:34.240 align:middle line:84% other thing than it would be to get a computer 00:40:34.240 --> 00:40:36.910 align:middle line:84% to figure out which thing c-commands which other things. 00:40:36.910 --> 00:40:38.603 align:middle line:84% That's a much more complicated task, 00:40:38.603 --> 00:40:41.020 align:middle line:84% it seems to me, not that I know anything about this stuff. 00:40:41.020 --> 00:40:43.840 align:middle line:84% And yet all of us go for the more complicated task. 00:40:43.840 --> 00:40:44.840 align:middle line:90% We're built that way. 00:40:44.840 --> 00:40:45.850 align:middle line:90% We can't help it. 00:40:45.850 --> 00:40:47.800 align:middle line:84% So language doesn't care which thing 00:40:47.800 --> 00:40:49.000 align:middle line:90% precedes which other thing. 00:40:49.000 --> 00:40:51.520 align:middle line:84% It cares which things c-commands which other thing, always, 00:40:51.520 --> 00:40:53.740 align:middle line:84% in every language, over and over again. 00:40:53.740 --> 00:40:54.490 align:middle line:90% Yeah. 00:40:54.490 --> 00:40:58.490 align:middle line:84% AUDIENCE: How do we draw a tree when the subordinate phrase is 00:40:58.490 --> 00:40:59.127 align:middle line:90% [INAUDIBLE]? 00:40:59.127 --> 00:40:59.960 align:middle line:90% NORVIN RICHARDS: Oh. 00:40:59.960 --> 00:41:02.540 align:middle line:84% Well, we could do different ways. 00:41:02.540 --> 00:41:06.710 align:middle line:90% 00:41:06.710 --> 00:41:08.510 align:middle line:90% Where will I put this? 00:41:08.510 --> 00:41:09.350 align:middle line:90% I'll put it up here. 00:41:09.350 --> 00:41:13.320 align:middle line:90% 00:41:13.320 --> 00:41:17.310 align:middle line:84% So maybe you would start by drawing the TP first. 00:41:17.310 --> 00:41:33.970 align:middle line:90% So "Susan read a book," right? 00:41:33.970 --> 00:41:36.220 align:middle line:90% Do that for "Susan read a book." 00:41:36.220 --> 00:41:40.480 align:middle line:84% And then we need somewhere to hang "while she was eating." 00:41:40.480 --> 00:41:43.780 align:middle line:84% And we need it to be some kind of adjunct. 00:41:43.780 --> 00:41:50.400 align:middle line:84% So it's not in the same kind of relation as anything else. 00:41:50.400 --> 00:41:51.400 align:middle line:90% What am I trying to say? 00:41:51.400 --> 00:41:52.780 align:middle line:90% It's not selected by anything. 00:41:52.780 --> 00:41:55.490 align:middle line:84% There's no particular reason for it to be anywhere. 00:41:55.490 --> 00:41:59.140 align:middle line:84% So there are various ways we could do that. 00:41:59.140 --> 00:42:02.980 align:middle line:84% Notice, for example, that if this were an embedded clause, 00:42:02.980 --> 00:42:07.850 align:middle line:90% so if it were "I think that"-- 00:42:07.850 --> 00:42:12.556 align:middle line:84% so here's a C. It's going to be a CP-- 00:42:12.556 --> 00:42:15.730 align:middle line:90% sorry, a CP. 00:42:15.730 --> 00:42:18.580 align:middle line:84% So if we were doing, "I think that Susan read a book," 00:42:18.580 --> 00:42:20.563 align:middle line:84% the embedded clause "that Susan read a book" 00:42:20.563 --> 00:42:21.730 align:middle line:90% would look like this, right? 00:42:21.730 --> 00:42:24.430 align:middle line:84% It would have a CP here where I can just connect these. 00:42:24.430 --> 00:42:27.040 align:middle line:84% We have a CP with a C that that's 00:42:27.040 --> 00:42:29.800 align:middle line:90% got this TP as its sister. 00:42:29.800 --> 00:42:32.920 align:middle line:84% And if we were going to add this embedded clause, 00:42:32.920 --> 00:42:35.410 align:middle line:84% this adjuncty clause to the beginning 00:42:35.410 --> 00:42:40.360 align:middle line:84% of the embedded clause, while she was eating, 00:42:40.360 --> 00:42:41.830 align:middle line:84% let's think about places to put it. 00:42:41.830 --> 00:42:44.740 align:middle line:84% If it's going to be before "Susan," we could ask, 00:42:44.740 --> 00:42:45.940 align:middle line:90% does it go here? 00:42:45.940 --> 00:42:47.710 align:middle line:90% Or does it go here? 00:42:47.710 --> 00:42:50.770 align:middle line:84% So "I think that, while she was eating, 00:42:50.770 --> 00:42:55.290 align:middle line:84% Susan read a book" is, I think, something that we can say. 00:42:55.290 --> 00:42:57.855 align:middle line:84% "I think, while she was eating, that Susan read a book." 00:42:57.855 --> 00:43:02.520 align:middle line:90% 00:43:02.520 --> 00:43:04.620 align:middle line:84% I think does not mean the same thing. 00:43:04.620 --> 00:43:06.820 align:middle line:84% There's something wrong with that. 00:43:06.820 --> 00:43:08.850 align:middle line:84% So this is at least one place that it can 00:43:08.850 --> 00:43:13.097 align:middle line:90% go between the C and the TP. 00:43:13.097 --> 00:43:15.180 align:middle line:84% And given that it can go between the C and the TP, 00:43:15.180 --> 00:43:17.370 align:middle line:84% maybe the easiest thing to do would be to just 00:43:17.370 --> 00:43:20.550 align:middle line:90% make it an adjunct on TP. 00:43:20.550 --> 00:43:22.980 align:middle line:90% We'll duplicate the TP node. 00:43:22.980 --> 00:43:27.030 align:middle line:84% And we'll have a CP here where I won't do the insides of it, 00:43:27.030 --> 00:43:29.395 align:middle line:84% but that's where you would put "while she was eating." 00:43:29.395 --> 00:43:30.270 align:middle line:90% Does that make sense? 00:43:30.270 --> 00:43:32.830 align:middle line:90% 00:43:32.830 --> 00:43:34.235 align:middle line:90% Yeah? 00:43:34.235 --> 00:43:34.735 align:middle line:90% OK. 00:43:34.735 --> 00:43:38.060 align:middle line:90% 00:43:38.060 --> 00:43:39.710 align:middle line:90% So that's binding theory. 00:43:39.710 --> 00:43:41.960 align:middle line:84% I think I flagged the fact there are many, many things 00:43:41.960 --> 00:43:43.127 align:middle line:90% to say about binding theory. 00:43:43.127 --> 00:43:44.630 align:middle line:90% That's part of it. 00:43:44.630 --> 00:43:47.780 align:middle line:84% Reasons to be happy about binding theory, 00:43:47.780 --> 00:43:50.060 align:middle line:84% I first showed you binding theory as a probe 00:43:50.060 --> 00:43:52.278 align:middle line:84% into c-command relations, but there's so much more. 00:43:52.278 --> 00:43:54.570 align:middle line:84% There are lots of things we can do with binding theory. 00:43:54.570 --> 00:43:56.300 align:middle line:90% Let me show you one. 00:43:56.300 --> 00:43:57.290 align:middle line:90% Here's a sentence. 00:43:57.290 --> 00:43:59.270 align:middle line:90% "Mary decided to leave." 00:43:59.270 --> 00:44:00.590 align:middle line:90% Yeah, fine sentence. 00:44:00.590 --> 00:44:03.400 align:middle line:90% 00:44:03.400 --> 00:44:05.515 align:middle line:84% How many TPs are in this sentence? 00:44:05.515 --> 00:44:09.360 align:middle line:90% 00:44:09.360 --> 00:44:12.720 align:middle line:90% well there's at least one. 00:44:12.720 --> 00:44:14.265 align:middle line:90% What's "to leave"? 00:44:14.265 --> 00:44:19.290 align:middle line:90% 00:44:19.290 --> 00:44:21.900 align:middle line:90% "To" isn't a preposition here. 00:44:21.900 --> 00:44:27.090 align:middle line:84% This isn't like "Mary went to Toledo," yeah? 00:44:27.090 --> 00:44:39.920 align:middle line:84% In fact, this "to" is in complementary distribution 00:44:39.920 --> 00:44:44.300 align:middle line:84% with a bunch of different things. 00:44:44.300 --> 00:44:46.070 align:middle line:90% You can't say, for example-- 00:44:46.070 --> 00:44:47.570 align:middle line:84% you can say "Mary decided to leave." 00:44:47.570 --> 00:44:49.910 align:middle line:84% You can't say "Mary decided will leave" 00:44:49.910 --> 00:44:57.650 align:middle line:84% or "should leave" or, for that matter, "Mary decided left." 00:44:57.650 --> 00:44:59.480 align:middle line:90% These are not sentences. 00:44:59.480 --> 00:45:05.720 align:middle line:84% So these are all bad, which makes people think that maybe 00:45:05.720 --> 00:45:09.860 align:middle line:84% "to" is the same kind of thing as "will" and "should." 00:45:09.860 --> 00:45:12.443 align:middle line:84% So here's a spot where you can only have one of these things. 00:45:12.443 --> 00:45:14.360 align:middle line:84% You can have "will," or you can have "should," 00:45:14.360 --> 00:45:17.170 align:middle line:90% or you can "have to." 00:45:17.170 --> 00:45:19.180 align:middle line:90% And so people think-- yeah. 00:45:19.180 --> 00:45:21.730 align:middle line:84% AUDIENCE: Is "to" not joined with the verb 00:45:21.730 --> 00:45:24.910 align:middle line:84% itself, though, as part of the infinitive? 00:45:24.910 --> 00:45:26.295 align:middle line:90% NORVIN RICHARDS: I don't know. 00:45:26.295 --> 00:45:28.420 align:middle line:84% This gets into questions about whether you believed 00:45:28.420 --> 00:45:33.040 align:middle line:84% your English teacher when she or he told you that you cannot say 00:45:33.040 --> 00:45:39.070 align:middle line:84% things like "Mary decided to immediately leave, 00:45:39.070 --> 00:45:41.935 align:middle line:84% but we decided to boldly go where no one has gone before." 00:45:41.935 --> 00:45:44.800 align:middle line:90% 00:45:44.800 --> 00:45:50.070 align:middle line:84% There are languages in which the marker of the infinitive 00:45:50.070 --> 00:45:51.820 align:middle line:84% is a morpheme that's attached to the verb. 00:45:51.820 --> 00:45:53.230 align:middle line:90% Yeah, you're absolutely right. 00:45:53.230 --> 00:45:55.390 align:middle line:84% But in English, it's not so clear that it is. 00:45:55.390 --> 00:45:58.240 align:middle line:84% There's stuff like adverbs that you can put between the "to" 00:45:58.240 --> 00:46:01.760 align:middle line:90% and the verb, at least I can. 00:46:01.760 --> 00:46:03.120 align:middle line:90% Yeah? 00:46:03.120 --> 00:46:04.090 align:middle line:90% All right, Joseph? 00:46:04.090 --> 00:46:07.370 align:middle line:84% AUDIENCE: Is the "to" shorthand for "that she should," "would," 00:46:07.370 --> 00:46:07.870 align:middle line:90% or-- 00:46:07.870 --> 00:46:09.840 align:middle line:84% NORVIN RICHARDS: Well, so this sentence 00:46:09.840 --> 00:46:11.590 align:middle line:84% means more or less the same thing as "Mary 00:46:11.590 --> 00:46:12.970 align:middle line:90% decided that she would leave." 00:46:12.970 --> 00:46:20.710 align:middle line:90% 00:46:20.710 --> 00:46:23.278 align:middle line:84% So interesting fact, you can't have "that" 00:46:23.278 --> 00:46:24.570 align:middle line:90% if you're going to have a "to." 00:46:24.570 --> 00:46:27.920 align:middle line:90% 00:46:27.920 --> 00:46:28.420 align:middle line:90% Yeah. 00:46:28.420 --> 00:46:30.100 align:middle line:84% If you have a "that," then you cannot have a "to." 00:46:30.100 --> 00:46:32.920 align:middle line:84% If you have "that," then you can have all these other things 00:46:32.920 --> 00:46:35.080 align:middle line:84% that would be here if "to" were not here. 00:46:35.080 --> 00:46:36.730 align:middle line:84% You could have a "will," or a "should," 00:46:36.730 --> 00:46:39.140 align:middle line:84% or a past tense, or all of these other things. 00:46:39.140 --> 00:46:40.630 align:middle line:84% That's the relation that we talked 00:46:40.630 --> 00:46:41.797 align:middle line:90% about in terms of selection. 00:46:41.797 --> 00:46:46.660 align:middle line:84% We said C selects T. T is selected by C. 00:46:46.660 --> 00:46:52.180 align:middle line:84% So these are all reasons to think maybe that "to" is a T. 00:46:52.180 --> 00:46:54.040 align:middle line:84% There are two TPs in this sentence. 00:46:54.040 --> 00:46:56.560 align:middle line:90% There's "Mary decided to leave." 00:46:56.560 --> 00:46:57.250 align:middle line:90% That's one TP. 00:46:57.250 --> 00:47:02.410 align:middle line:84% And that "to leave" is another TP, OK? 00:47:02.410 --> 00:47:04.090 align:middle line:90% Are you buying this? 00:47:04.090 --> 00:47:04.930 align:middle line:90% You should. 00:47:04.930 --> 00:47:07.080 align:middle line:90% Yeah. 00:47:07.080 --> 00:47:09.000 align:middle line:90% OK, so maybe there are two TPs. 00:47:09.000 --> 00:47:11.713 align:middle line:84% But if there are two TPs in this sentence, 00:47:11.713 --> 00:47:14.130 align:middle line:84% we have this principle, the extended projection principle, 00:47:14.130 --> 00:47:17.460 align:middle line:84% that says every TP must have a specifier. 00:47:17.460 --> 00:47:22.230 align:middle line:84% But we only seem to have one subject, "Mary," 00:47:22.230 --> 00:47:25.240 align:middle line:90% who's doing the deciding. 00:47:25.240 --> 00:47:29.080 align:middle line:84% Now, there's a classic answer to this question, which 00:47:29.080 --> 00:47:35.350 align:middle line:84% says English is a language with many pronouns, and one of them 00:47:35.350 --> 00:47:38.050 align:middle line:90% is invisible. 00:47:38.050 --> 00:47:41.770 align:middle line:84% And the invisible one is the one that is the subject of "leave." 00:47:41.770 --> 00:47:44.260 align:middle line:84% It's the thing that's satisfying the EPP here. 00:47:44.260 --> 00:47:48.400 align:middle line:84% So it isn't that TP doesn't have a specifier in "to leave." 00:47:48.400 --> 00:47:49.960 align:middle line:90% That TP does have a specifier. 00:47:49.960 --> 00:47:54.070 align:middle line:84% It's just an invisible specifier. 00:47:54.070 --> 00:47:59.480 align:middle line:84% When this theory was first offered, 00:47:59.480 --> 00:48:02.740 align:middle line:84% people reacted the way you might expect them to, 00:48:02.740 --> 00:48:04.850 align:middle line:84% sort of the way some of you possibly mentally are: 00:48:04.850 --> 00:48:08.320 align:middle line:84% "I mean, come on, you're just trying 00:48:08.320 --> 00:48:10.750 align:middle line:84% to avoid having to say that the EPP is not right 00:48:10.750 --> 00:48:14.560 align:middle line:84% or parameterize the EPP to different kinds of positions." 00:48:14.560 --> 00:48:16.120 align:middle line:84% But actually, binding theory gives us 00:48:16.120 --> 00:48:19.270 align:middle line:84% some reasons to believe that there is such a thing as PRO. 00:48:19.270 --> 00:48:22.040 align:middle line:84% And let me show you some of the arguments. 00:48:22.040 --> 00:48:24.670 align:middle line:84% So in this case, that's a story that says, 00:48:24.670 --> 00:48:27.010 align:middle line:84% no, the EPP is being satisfied in the embedded clause 00:48:27.010 --> 00:48:30.085 align:middle line:84% by an invisible pronoun that refers, in this case, to Mary. 00:48:30.085 --> 00:48:33.120 align:middle line:90% 00:48:33.120 --> 00:48:36.000 align:middle line:90% Here are some data. 00:48:36.000 --> 00:48:37.920 align:middle line:84% "John promised Mary to defend himself." 00:48:37.920 --> 00:48:39.900 align:middle line:84% "John promised Mary to defend herself." 00:48:39.900 --> 00:48:41.670 align:middle line:84% "John told Mary to defend himself." 00:48:41.670 --> 00:48:44.310 align:middle line:84% "John told Mary to defend herself"-- 00:48:44.310 --> 00:48:47.910 align:middle line:84% sentences with grammaticality judgments. 00:48:47.910 --> 00:48:50.110 align:middle line:84% I'm going to copy these sentences onto the board 00:48:50.110 --> 00:48:52.360 align:middle line:84% because I want to do various graphic things with them. 00:48:52.360 --> 00:48:54.450 align:middle line:84% And while I'm copying them, I'll ask 00:48:54.450 --> 00:48:58.740 align:middle line:84% you to look at those sentences and ask yourself whether you 00:48:58.740 --> 00:49:00.300 align:middle line:90% get these judgments. 00:49:00.300 --> 00:49:11.590 align:middle line:84% So "John promised Mary to defend herself"-- 00:49:11.590 --> 00:49:13.902 align:middle line:90% "himself." 00:49:13.902 --> 00:49:15.360 align:middle line:84% Did I put those in the wrong order? 00:49:15.360 --> 00:49:16.050 align:middle line:90% Yes, I did. 00:49:16.050 --> 00:49:19.620 align:middle line:90% 00:49:19.620 --> 00:49:33.492 align:middle line:84% And then "John told Mary to defend himself"-- 00:49:33.492 --> 00:49:40.110 align:middle line:84% "herself," where the middle two-- so this is bad 00:49:40.110 --> 00:49:41.390 align:middle line:90% and this is bad. 00:49:41.390 --> 00:49:42.640 align:middle line:90% Do people get these judgments? 00:49:42.640 --> 00:49:43.420 align:middle line:90% Is this true? 00:49:43.420 --> 00:49:45.050 align:middle line:90% Am I just making this up? 00:49:45.050 --> 00:49:46.150 align:middle line:90% Nobody wants to protest? 00:49:46.150 --> 00:49:46.690 align:middle line:90% OK. 00:49:46.690 --> 00:49:51.250 align:middle line:84% So I told you before we're going to use binding theory 00:49:51.250 --> 00:49:56.353 align:middle line:84% to find things out about how far away things are, which things 00:49:56.353 --> 00:49:58.270 align:middle line:84% c-command which other things, now that we have 00:49:58.270 --> 00:50:02.800 align:middle line:84% this theory that we think, we hope, is more or less reliable. 00:50:02.800 --> 00:50:04.300 align:middle line:84% We know, when we look at an anaphor, 00:50:04.300 --> 00:50:06.758 align:middle line:84% it had better have an antecedent that's close enough to it. 00:50:06.758 --> 00:50:09.220 align:middle line:84% And we think that close enough is more or less 00:50:09.220 --> 00:50:12.950 align:middle line:90% within the same TP. 00:50:12.950 --> 00:50:14.390 align:middle line:84% Forget it, now, that I've done all 00:50:14.390 --> 00:50:15.557 align:middle line:90% that talking about anaphors. 00:50:15.557 --> 00:50:17.580 align:middle line:84% Forget about the anaphors for a second. 00:50:17.580 --> 00:50:25.895 align:middle line:84% If I say "John promised Mary to leave," who's going to leave? 00:50:25.895 --> 00:50:26.520 align:middle line:90% AUDIENCE: John. 00:50:26.520 --> 00:50:27.540 align:middle line:90% NORVIN RICHARDS: John. 00:50:27.540 --> 00:50:32.135 align:middle line:84% And if I say "John told Mary to leave," who's going to leave? 00:50:32.135 --> 00:50:32.760 align:middle line:90% AUDIENCE: Mary. 00:50:32.760 --> 00:50:33.930 align:middle line:90% NORVIN RICHARDS: Mary, yeah? 00:50:33.930 --> 00:50:38.370 align:middle line:84% So the understood subjects of these embedded clauses, 00:50:38.370 --> 00:50:41.400 align:middle line:84% the PRO if there's a PRO, is going 00:50:41.400 --> 00:50:49.590 align:middle line:84% to refer to John when the first verb is promised. 00:50:49.590 --> 00:50:53.730 align:middle line:84% But it's going to refer to Mary when-- 00:50:53.730 --> 00:50:56.940 align:middle line:84% I'm just going to give it a subscript, Mary-- 00:50:56.940 --> 00:50:59.880 align:middle line:90% the first verb is told, yeah? 00:50:59.880 --> 00:51:02.580 align:middle line:84% So let me do this the easy way first. 00:51:02.580 --> 00:51:07.140 align:middle line:84% If there is such a thing as PRO, we ask ourself, these anaphors, 00:51:07.140 --> 00:51:10.050 align:middle line:84% they need to get bound within the smallest 00:51:10.050 --> 00:51:12.960 align:middle line:90% TP that contains the anaphor. 00:51:12.960 --> 00:51:15.330 align:middle line:84% And the smallest TP that contains the anaphor is 00:51:15.330 --> 00:51:19.820 align:middle line:90% going to be this, yeah? 00:51:19.820 --> 00:51:24.410 align:middle line:84% And so when we ask ourself "Is the anaphor 00:51:24.410 --> 00:51:28.230 align:middle line:84% bound by a noun phrase that's close enough to it?" 00:51:28.230 --> 00:51:30.410 align:middle line:90% well, the answer is yes here. 00:51:30.410 --> 00:51:33.560 align:middle line:84% Because "himself" gets to be bound by that PRO, "John." 00:51:33.560 --> 00:51:34.910 align:middle line:90% That refers to John. 00:51:34.910 --> 00:51:37.260 align:middle line:90% And so "himself" can be John. 00:51:37.260 --> 00:51:39.260 align:middle line:84% It can't be "herself" because the PRO is "John." 00:51:39.260 --> 00:51:43.700 align:middle line:84% It's not "Mary," whereas, when the higher verb is "told," 00:51:43.700 --> 00:51:45.140 align:middle line:90% then the PRO is "Mary." 00:51:45.140 --> 00:51:47.090 align:middle line:84% And so now "himself" is bad because "himself" 00:51:47.090 --> 00:51:47.990 align:middle line:90% can't refer to Mary. 00:51:47.990 --> 00:51:49.370 align:middle line:90% And "herself" is good. 00:51:49.370 --> 00:51:51.380 align:middle line:90% It can get bound by "Mary." 00:51:51.380 --> 00:51:53.330 align:middle line:84% So we can capture all of these facts 00:51:53.330 --> 00:51:55.700 align:middle line:84% if we're willing to posit the PRO, which 00:51:55.700 --> 00:51:57.410 align:middle line:84% refers to who it should refer to, 00:51:57.410 --> 00:52:00.800 align:middle line:84% the person who's going to do the defending. 00:52:00.800 --> 00:52:03.110 align:middle line:84% That'll depend on what the higher verb is. 00:52:03.110 --> 00:52:05.240 align:middle line:84% Interesting fact, which people work on-- what 00:52:05.240 --> 00:52:07.220 align:middle line:90% determines who PRO refers to? 00:52:07.220 --> 00:52:11.390 align:middle line:84% Why is it "John" for "promise," but "Mary" for "defend"? 00:52:11.390 --> 00:52:14.990 align:middle line:84% It's what's called subject control and object control. 00:52:14.990 --> 00:52:16.940 align:middle line:84% But once we believe in PRO, we have 00:52:16.940 --> 00:52:20.690 align:middle line:84% a handle on why the anaphors refer to who they refer to. 00:52:20.690 --> 00:52:22.340 align:middle line:90% OK? 00:52:22.340 --> 00:52:23.690 align:middle line:90% Now, let's do this the hard way. 00:52:23.690 --> 00:52:25.250 align:middle line:84% What if we didn't believe in PRO? 00:52:25.250 --> 00:52:28.690 align:middle line:90% 00:52:28.690 --> 00:52:36.330 align:middle line:84% Well, if we don't believe in PRO, woo, what we want 00:52:36.330 --> 00:52:44.550 align:middle line:84% is for this position to be bindable by "John," 00:52:44.550 --> 00:52:45.780 align:middle line:90% but not by "Mary." 00:52:45.780 --> 00:52:49.890 align:middle line:90% 00:52:49.890 --> 00:52:54.750 align:middle line:84% And we want this position to be bindable by "Mary," 00:52:54.750 --> 00:52:55.620 align:middle line:90% but not by "John." 00:52:55.620 --> 00:52:58.530 align:middle line:90% 00:52:58.530 --> 00:53:01.765 align:middle line:84% And this is going to be hard, right? 00:53:01.765 --> 00:53:03.640 align:middle line:84% The first one is going to be especially hard. 00:53:03.640 --> 00:53:05.880 align:middle line:84% So far, our conditions just say, anaphors 00:53:05.880 --> 00:53:07.380 align:middle line:84% have to get c-commanded by something 00:53:07.380 --> 00:53:11.940 align:middle line:84% that's close enough to them and corefers with them. 00:53:11.940 --> 00:53:14.313 align:middle line:84% But if "John" is close enough to that anaphor position, 00:53:14.313 --> 00:53:15.480 align:middle line:90% well, then surely "Mary" is. 00:53:15.480 --> 00:53:16.890 align:middle line:90% "Mary" is closer. 00:53:16.890 --> 00:53:19.500 align:middle line:90% There's no way to do that. 00:53:19.500 --> 00:53:22.830 align:middle line:84% I'll have to come up with something very complicated. 00:53:22.830 --> 00:53:24.690 align:middle line:84% And the second set of examples is also 00:53:24.690 --> 00:53:30.660 align:middle line:84% going to be kind of hard because "Mary" is the object of "told." 00:53:30.660 --> 00:53:33.660 align:middle line:84% So this anaphor will need to have a binding domain that 00:53:33.660 --> 00:53:36.900 align:middle line:84% gets up to the higher verb phrase, a domain in which 00:53:36.900 --> 00:53:40.840 align:middle line:90% its anaphor finds a binder. 00:53:40.840 --> 00:53:42.840 align:middle line:84% It's going to need to find a binder in a domain. 00:53:42.840 --> 00:53:45.630 align:middle line:84% We have to say, OK, we were wrong about TPs. 00:53:45.630 --> 00:53:48.510 align:middle line:84% It's, like, the next verb phrase or something. 00:53:48.510 --> 00:53:50.610 align:middle line:84% The domain in which it finds a binder 00:53:50.610 --> 00:53:53.845 align:middle line:84% has to be this new special kind of domain that doesn't seem 00:53:53.845 --> 00:53:54.970 align:middle line:90% to be useful anywhere else. 00:53:54.970 --> 00:53:57.060 align:middle line:90% It's a mess, basically. 00:53:57.060 --> 00:53:59.640 align:middle line:84% And PRO gets us out of the mess if we're 00:53:59.640 --> 00:54:04.840 align:middle line:84% willing to put PRO in the places where we want to put it. 00:54:04.840 --> 00:54:08.790 align:middle line:84% So one reason to be happy about binding theory 00:54:08.790 --> 00:54:11.940 align:middle line:84% is that it gives us a reason to make a move that we might 00:54:11.940 --> 00:54:13.500 align:middle line:90% have wanted to make anyway. 00:54:13.500 --> 00:54:16.950 align:middle line:84% These infinitives are acting like they 00:54:16.950 --> 00:54:19.530 align:middle line:90% have subjects in two respects. 00:54:19.530 --> 00:54:22.470 align:middle line:84% They are satisfying EPP, first of all. 00:54:22.470 --> 00:54:25.980 align:middle line:84% And they are acting like they have a subject that 00:54:25.980 --> 00:54:27.840 align:middle line:84% corefers with somebody in the higher clause, 00:54:27.840 --> 00:54:31.110 align:middle line:84% either "John" or "Mary," for purposes of binding theory. 00:54:31.110 --> 00:54:35.580 align:middle line:84% Binding theory is simplest if there is a pronoun there. 00:54:35.580 --> 00:54:40.510 align:middle line:84% Now, minor problem, we can't see a pronoun there. 00:54:40.510 --> 00:54:44.422 align:middle line:84% But if we're willing to rely on binding theory a little bit, 00:54:44.422 --> 00:54:46.380 align:middle line:84% we can use binding theory to convince ourselves 00:54:46.380 --> 00:54:47.910 align:middle line:84% that the pronoun actually is there. 00:54:47.910 --> 00:54:49.440 align:middle line:90% It's just that we can't see it. 00:54:49.440 --> 00:54:52.140 align:middle line:84% We have a new detector for pronouns. 00:54:52.140 --> 00:54:54.185 align:middle line:90% We can find them this way. 00:54:54.185 --> 00:54:55.060 align:middle line:90% Does that make sense? 00:54:55.060 --> 00:54:56.500 align:middle line:84% Are all of you appropriately skeptical? 00:54:56.500 --> 00:54:58.120 align:middle line:84% Is anybody inappropriately skeptical? 00:54:58.120 --> 00:55:01.315 align:middle line:84% Do you have questions about any of this chain of reasoning? 00:55:01.315 --> 00:55:03.370 align:middle line:84% It's a reason to take seriously the idea 00:55:03.370 --> 00:55:06.280 align:middle line:84% that there are invisible pronouns sometimes. 00:55:06.280 --> 00:55:08.710 align:middle line:84% We talked about invisible pronouns before, I think. 00:55:08.710 --> 00:55:11.860 align:middle line:84% I think I gave you examples like I can 00:55:11.860 --> 00:55:18.300 align:middle line:90% say "Defend yourself," yeah? 00:55:18.300 --> 00:55:20.488 align:middle line:90% 00:55:20.488 --> 00:55:22.780 align:middle line:84% I think somebody said that they were taught in school-- 00:55:22.780 --> 00:55:24.960 align:middle line:84% I can't remember who it was-- that the subject 00:55:24.960 --> 00:55:28.080 align:middle line:84% of an imperative is understood as "you," that this 00:55:28.080 --> 00:55:29.760 align:middle line:90% means "You defend yourself." 00:55:29.760 --> 00:55:31.320 align:middle line:84% And one reason to take that seriously 00:55:31.320 --> 00:55:33.510 align:middle line:84% is that, yeah, an anaphor here is 00:55:33.510 --> 00:55:35.220 align:middle line:84% acting like there is a subject around, 00:55:35.220 --> 00:55:38.340 align:middle line:84% or at least another noun phrase around, that refers to "you." 00:55:38.340 --> 00:55:40.380 align:middle line:84% Because "Defend yourself" is fine. 00:55:40.380 --> 00:55:44.127 align:middle line:84% But "Defend myself"-- no good, right? 00:55:44.127 --> 00:55:45.960 align:middle line:84% No matter how much you might want to say it, 00:55:45.960 --> 00:55:47.168 align:middle line:90% that's not the way to say it. 00:55:47.168 --> 00:55:49.110 align:middle line:90% You say "Defend me," no? 00:55:49.110 --> 00:55:50.780 align:middle line:84% It's acting like the subject is "you." 00:55:50.780 --> 00:55:52.043 align:middle line:90% OK, you can't hear it. 00:55:52.043 --> 00:55:54.210 align:middle line:84% You can't hear the "you," but there's a "you" there. 00:55:54.210 --> 00:55:57.120 align:middle line:84% You can use these anaphors to find it. 00:55:57.120 --> 00:55:59.370 align:middle line:84% Same thing here-- in "John promised 00:55:59.370 --> 00:56:01.770 align:middle line:84% Mary to defend himself," there's a pronoun 00:56:01.770 --> 00:56:03.060 align:middle line:90% in the subject of "defend." 00:56:03.060 --> 00:56:07.740 align:middle line:84% It's really "John promised Mary him to defend himself." 00:56:07.740 --> 00:56:11.280 align:middle line:84% And that "him," that PRO, is referring back to "John." 00:56:11.280 --> 00:56:14.670 align:middle line:84% In "John told Mary her to defend herself," 00:56:14.670 --> 00:56:18.310 align:middle line:84% that "her" is referring back to "Mary." 00:56:18.310 --> 00:56:21.842 align:middle line:90% Binding theory tells us so. 00:56:21.842 --> 00:56:25.780 align:middle line:84% Yeah, yeah, yeah-- slide after slide after slide. 00:56:25.780 --> 00:56:28.210 align:middle line:90% Yes. 00:56:28.210 --> 00:56:31.240 align:middle line:84% OK, here's another reason to be happy about binding theory, 00:56:31.240 --> 00:56:33.370 align:middle line:84% another thing that it teaches us, 00:56:33.370 --> 00:56:35.770 align:middle line:84% a new argument for something that I showed you 00:56:35.770 --> 00:56:37.450 align:middle line:90% an argument for before. 00:56:37.450 --> 00:56:40.270 align:middle line:84% And now, I will show you a new argument, possibly 00:56:40.270 --> 00:56:41.080 align:middle line:90% a simpler argument. 00:56:41.080 --> 00:56:43.790 align:middle line:84% Actually, I should start with this one. 00:56:43.790 --> 00:56:45.970 align:middle line:84% "Which picture of himself did John like best"? 00:56:45.970 --> 00:56:49.090 align:middle line:90% First-- setup for the argument. 00:56:49.090 --> 00:56:53.140 align:middle line:84% Observation, "John" sure doesn't look 00:56:53.140 --> 00:56:56.600 align:middle line:90% like it c-commands himself. 00:56:56.600 --> 00:56:59.660 align:middle line:84% But "John" used to c-command "himself." 00:56:59.660 --> 00:57:05.040 align:middle line:84% And apparently, principle A can be satisfied with nostalgia. 00:57:05.040 --> 00:57:07.400 align:middle line:84% So it thinks back to the happy days 00:57:07.400 --> 00:57:10.190 align:middle line:84% when "John" used to c-command "himself," back before you 00:57:10.190 --> 00:57:11.870 align:middle line:90% did wh- movement. 00:57:11.870 --> 00:57:13.700 align:middle line:90% And it's satisfied. 00:57:13.700 --> 00:57:15.850 align:middle line:84% Principle A, it's easy going that way. 00:57:15.850 --> 00:57:18.320 align:middle line:84% So this starts off with "John liked 00:57:18.320 --> 00:57:19.958 align:middle line:90% which picture of himself best?" 00:57:19.958 --> 00:57:21.500 align:middle line:84% "Which picture of himself" starts off 00:57:21.500 --> 00:57:22.760 align:middle line:90% as the object of "like." 00:57:22.760 --> 00:57:25.550 align:middle line:90% And it wh- moves. 00:57:25.550 --> 00:57:26.870 align:middle line:90% It becomes the specifier of CP. 00:57:26.870 --> 00:57:29.350 align:middle line:84% Shall I draw a tree for that, or is that clear? 00:57:29.350 --> 00:57:29.850 align:middle line:90% I'm sorry. 00:57:29.850 --> 00:57:32.140 align:middle line:84% Those were yes/no questions with opposite answers. 00:57:32.140 --> 00:57:35.300 align:middle line:84% Who would like me to draw a tree for that? 00:57:35.300 --> 00:57:35.800 align:middle line:90% Yes? 00:57:35.800 --> 00:57:38.380 align:middle line:90% OK, let me draw a tree for that. 00:57:38.380 --> 00:57:41.410 align:middle line:90% I'll hide these trees. 00:57:41.410 --> 00:57:46.090 align:middle line:90% So-- stamp. 00:57:46.090 --> 00:57:51.520 align:middle line:90% 00:57:51.520 --> 00:57:52.680 align:middle line:90% John-- yeah. 00:57:52.680 --> 00:57:57.525 align:middle line:90% 00:57:57.525 --> 00:57:58.400 align:middle line:90% Here's a noun phrase. 00:57:58.400 --> 00:58:01.890 align:middle line:84% I'm going to leave out "best" just because we don't need it. 00:58:01.890 --> 00:58:05.570 align:middle line:84% Here's your DP, "which," and N bar, 00:58:05.570 --> 00:58:11.583 align:middle line:84% "picture," prepositional phrase, "of," 00:58:11.583 --> 00:58:12.875 align:middle line:90% and the noun phrase, "himself." 00:58:12.875 --> 00:58:15.490 align:middle line:90% 00:58:15.490 --> 00:58:18.530 align:middle line:84% OK, so this is just "John liked which picture of himself." 00:58:18.530 --> 00:58:19.780 align:middle line:90% We could put a "best" in here. 00:58:19.780 --> 00:58:20.860 align:middle line:84% I don't know why I left out "best." 00:58:20.860 --> 00:58:21.700 align:middle line:90% We'll put it here. 00:58:21.700 --> 00:58:31.020 align:middle line:90% 00:58:31.020 --> 00:58:34.070 align:middle line:84% There, so "John liked which picture of himself best?" 00:58:34.070 --> 00:58:37.580 align:middle line:84% And then wh- movement will take this noun phrase 00:58:37.580 --> 00:58:42.140 align:middle line:84% and move it to make it the specifier of a CP, which 00:58:42.140 --> 00:58:44.370 align:middle line:90% is up here. 00:58:44.370 --> 00:58:49.890 align:middle line:84% So now, we have "which picture of himself" 00:58:49.890 --> 00:58:51.690 align:middle line:90% in the specifier of CP. 00:58:51.690 --> 00:58:56.893 align:middle line:84% And "did" ends up here because we've moved T to C 00:58:56.893 --> 00:58:58.560 align:middle line:84% and done do- support, which is something 00:58:58.560 --> 00:59:00.390 align:middle line:90% that Faith made me talk about. 00:59:00.390 --> 00:59:02.610 align:middle line:90% Yeah? 00:59:02.610 --> 00:59:05.557 align:middle line:84% So "Which picture of himself did John like best?" 00:59:05.557 --> 00:59:07.140 align:middle line:84% And "which picture of himself"-- which 00:59:07.140 --> 00:59:09.270 align:middle line:84% I'd spelled without the final H, just to 00:59:09.270 --> 00:59:12.168 align:middle line:84% give you an additional challenge-- 00:59:12.168 --> 00:59:13.710 align:middle line:84% "which picture of himself" starts off 00:59:13.710 --> 00:59:14.790 align:middle line:90% as the sister of "like." 00:59:14.790 --> 00:59:17.770 align:middle line:84% And wh- moves and becomes the specifier of CP. 00:59:17.770 --> 00:59:18.570 align:middle line:90% Yeah? 00:59:18.570 --> 00:59:19.890 align:middle line:90% That's the derivation for that. 00:59:19.890 --> 00:59:23.490 align:middle line:90% 00:59:23.490 --> 00:59:28.510 align:middle line:84% So fact, which we're going to have to get used to-- 00:59:28.510 --> 00:59:30.910 align:middle line:84% after we do the movement, "John" no longer c-commands 00:59:30.910 --> 00:59:31.690 align:middle line:90% "himself," right? 00:59:31.690 --> 00:59:32.815 align:middle line:90% What does "John" c-command? 00:59:32.815 --> 00:59:35.890 align:middle line:84% Well, we ask which nodes dominate "John." 00:59:35.890 --> 00:59:37.250 align:middle line:90% And it's these nodes. 00:59:37.250 --> 00:59:39.545 align:middle line:84% So "John" c-commands its sister and everything 00:59:39.545 --> 00:59:40.420 align:middle line:90% its sister dominates. 00:59:40.420 --> 00:59:43.270 align:middle line:84% So "John" c-commands "like which picture of himself best." 00:59:43.270 --> 00:59:45.770 align:middle line:84% But "which picture of himself" has moved out of that domain. 00:59:45.770 --> 00:59:46.330 align:middle line:90% It's up here. 00:59:46.330 --> 00:59:49.460 align:middle line:90% 00:59:49.460 --> 00:59:54.280 align:middle line:84% Apparently, the fact that "himself" 00:59:54.280 --> 00:59:55.990 align:middle line:84% used to be down here in a position 00:59:55.990 --> 00:59:59.513 align:middle line:84% that "John" could c-command is good enough. 00:59:59.513 --> 01:00:01.180 align:middle line:84% It's a phenomenon called reconstruction. 01:00:01.180 --> 01:00:05.770 align:middle line:84% It's as though, when you are evaluating things 01:00:05.770 --> 01:00:08.340 align:middle line:84% for binding theory, you get to rewind the tape a little bit 01:00:08.340 --> 01:00:09.430 align:middle line:90% and think, oh, yeah. 01:00:09.430 --> 01:00:10.513 align:middle line:90% This used to be over here. 01:00:10.513 --> 01:00:11.590 align:middle line:90% That's good enough. 01:00:11.590 --> 01:00:15.820 align:middle line:84% Various ways to talk about that, but that's one. 01:00:15.820 --> 01:00:18.490 align:middle line:90% So file that fact. 01:00:18.490 --> 01:00:19.480 align:middle line:90% Yes, I just said that. 01:00:19.480 --> 01:00:19.980 align:middle line:90% Thank you. 01:00:19.980 --> 01:00:21.760 align:middle line:84% This phenomenon is called reconstruction. 01:00:21.760 --> 01:00:25.700 align:middle line:84% You treat something which is moved as though it hadn't. 01:00:25.700 --> 01:00:26.480 align:middle line:90% OK. 01:00:26.480 --> 01:00:29.900 align:middle line:84% Now, let's think about longer, more complicated questions, 01:00:29.900 --> 01:00:33.430 align:middle line:84% like "Which picture did John think that Mary bought?" 01:00:33.430 --> 01:00:37.810 align:middle line:84% So "which picture" has wh- moved, right? 01:00:37.810 --> 01:00:41.410 align:middle line:84% But we talked about this question about whether it 01:00:41.410 --> 01:00:44.710 align:middle line:84% has moved once in one mighty leap, 01:00:44.710 --> 01:00:46.480 align:middle line:84% started off as the object of "by" 01:00:46.480 --> 01:00:48.760 align:middle line:84% and moved through all of that intervening material 01:00:48.760 --> 01:00:51.760 align:middle line:84% and became the specifier of the higher CP, 01:00:51.760 --> 01:00:54.880 align:middle line:90% or moved in a series of hops. 01:00:54.880 --> 01:00:57.940 align:middle line:84% Notice that "John thinks that Mary bought a picture" 01:00:57.940 --> 01:00:59.560 align:middle line:90% has two CPs in it. 01:00:59.560 --> 01:01:01.420 align:middle line:84% There's "that Mary bought a picture." 01:01:01.420 --> 01:01:02.710 align:middle line:90% That's the embedded clause. 01:01:02.710 --> 01:01:05.350 align:middle line:84% And then there's the matrix clause. 01:01:05.350 --> 01:01:07.505 align:middle line:90% And we raised the question. 01:01:07.505 --> 01:01:09.130 align:middle line:84% And I gave you all of these Dinka facts 01:01:09.130 --> 01:01:12.610 align:middle line:84% that were meant to get you to take seriously the possibility 01:01:12.610 --> 01:01:15.550 align:middle line:84% that the second picture is the right one, 01:01:15.550 --> 01:01:17.890 align:middle line:84% that movement is what is called successive cyclic. 01:01:17.890 --> 01:01:19.570 align:middle line:90% That is you move in short hops. 01:01:19.570 --> 01:01:21.670 align:middle line:90% You don't move in mighty leaps. 01:01:21.670 --> 01:01:25.630 align:middle line:84% So a wh- phrase, it's not just that it likes 01:01:25.630 --> 01:01:27.190 align:middle line:90% to move to the specifier of CP. 01:01:27.190 --> 01:01:29.350 align:middle line:84% It likes to land in every specifier of CP 01:01:29.350 --> 01:01:30.940 align:middle line:90% that it can find along the way. 01:01:30.940 --> 01:01:32.650 align:middle line:84% It moves in this series of short hops. 01:01:32.650 --> 01:01:34.150 align:middle line:84% And I gave you some Dinka facts that 01:01:34.150 --> 01:01:37.600 align:middle line:84% were meant to get you to take that possibility seriously. 01:01:37.600 --> 01:01:40.540 align:middle line:84% Here's another set of facts that might also 01:01:40.540 --> 01:01:42.850 align:middle line:84% get you to take that possibility seriously. 01:01:42.850 --> 01:01:45.370 align:middle line:84% Think about a sentence like, "Which picture of himself 01:01:45.370 --> 01:01:48.640 align:middle line:84% did John think that Mary bought?" 01:01:48.640 --> 01:01:50.080 align:middle line:90% First of all, it's fine. 01:01:50.080 --> 01:01:51.220 align:middle line:90% Do people agree? 01:01:51.220 --> 01:01:53.660 align:middle line:84% So "himself" can refer to "John." 01:01:53.660 --> 01:01:55.660 align:middle line:84% All right, now, let's think about why it's fine. 01:01:55.660 --> 01:01:58.600 align:middle line:90% 01:01:58.600 --> 01:02:00.730 align:middle line:84% "Which picture of himself" starts off 01:02:00.730 --> 01:02:03.990 align:middle line:84% as the sister of "by," starts off in the embedded clause. 01:02:03.990 --> 01:02:06.490 align:middle line:84% And it ends up, well, in the position where it's pronounced, 01:02:06.490 --> 01:02:07.990 align:middle line:90% in the matrix clause. 01:02:07.990 --> 01:02:10.330 align:middle line:84% Is "himself" bound by "John" in the position 01:02:10.330 --> 01:02:12.940 align:middle line:90% where it's pronounced? 01:02:12.940 --> 01:02:16.090 align:middle line:90% The specifier of the highest TP? 01:02:16.090 --> 01:02:17.635 align:middle line:84% Does "John" c-command that position? 01:02:17.635 --> 01:02:20.860 align:middle line:90% 01:02:20.860 --> 01:02:21.940 align:middle line:90% No, right? 01:02:21.940 --> 01:02:23.260 align:middle line:90% Shall I draw a tree for this? 01:02:23.260 --> 01:02:26.020 align:middle line:84% Would that be a good thing to do? 01:02:26.020 --> 01:02:27.932 align:middle line:90% Let's draw a tree. 01:02:27.932 --> 01:02:30.390 align:middle line:84% "Which picture of himself did John think that Mary bought?" 01:02:30.390 --> 01:02:34.930 align:middle line:90% 01:02:34.930 --> 01:02:35.545 align:middle line:90% Here's "John." 01:02:35.545 --> 01:02:40.140 align:middle line:90% 01:02:40.140 --> 01:02:44.770 align:middle line:90% "Think," there's the CP. 01:02:44.770 --> 01:02:48.340 align:middle line:90% "That," there's a TP. 01:02:48.340 --> 01:02:54.880 align:middle line:84% "Mary," here's T. And here's a VP and then "bought." 01:02:54.880 --> 01:02:56.980 align:middle line:84% And so we start off with "which picture," 01:02:56.980 --> 01:02:59.563 align:middle line:84% which I'm running out of room, so I'm going to use a triangle. 01:02:59.563 --> 01:03:00.910 align:middle line:90% Here's "which picture." 01:03:00.910 --> 01:03:02.410 align:middle line:84% I'm going to use a triangle, and I'm 01:03:02.410 --> 01:03:04.478 align:middle line:84% going to abbreviate picture as "pic." 01:03:04.478 --> 01:03:05.770 align:middle line:90% And the question we're asking-- 01:03:05.770 --> 01:03:09.710 align:middle line:84% and so here's the tree we start off with, yeah? 01:03:09.710 --> 01:03:12.140 align:middle line:84% Anybody wish to object to any aspects of this tree? 01:03:12.140 --> 01:03:13.400 align:middle line:90% Is this is OK? 01:03:13.400 --> 01:03:14.780 align:middle line:90% All right. 01:03:14.780 --> 01:03:17.480 align:middle line:84% And the question we're asking ourselves is, this noun phrase, 01:03:17.480 --> 01:03:21.210 align:middle line:90% does it move just once to here? 01:03:21.210 --> 01:03:25.680 align:middle line:84% So we have which picture of himself. 01:03:25.680 --> 01:03:26.180 align:middle line:90% [SNEEZE] 01:03:26.180 --> 01:03:28.885 align:middle line:90% Bless you. 01:03:28.885 --> 01:03:33.000 align:middle line:84% Does it move just once, or does it stop along the way here? 01:03:33.000 --> 01:03:34.820 align:middle line:84% That's the question we're asking ourselves. 01:03:34.820 --> 01:03:36.570 align:middle line:84% And now, we're going to use binding theory 01:03:36.570 --> 01:03:39.480 align:middle line:90% to come up with the answer. 01:03:39.480 --> 01:03:42.780 align:middle line:84% Does "John" c-command "himself" here? 01:03:42.780 --> 01:03:43.950 align:middle line:90% No. 01:03:43.950 --> 01:03:45.660 align:middle line:84% I just said "no" before I drew the tree. 01:03:45.660 --> 01:03:47.077 align:middle line:84% But now, that I've drawn the tree, 01:03:47.077 --> 01:03:49.110 align:middle line:84% you can see that the answer is, in fact, no. 01:03:49.110 --> 01:03:52.020 align:middle line:84% So "John" doesn't c-command this position. 01:03:52.020 --> 01:03:56.240 align:middle line:84% Does "John" c-command "himself" when himself is down here? 01:03:56.240 --> 01:03:57.410 align:middle line:90% Yeah, right? 01:03:57.410 --> 01:03:59.160 align:middle line:84% So what's the c-command domain for "John"? 01:03:59.160 --> 01:04:01.860 align:middle line:84% Well, it's its sister, that T bar, this T bar here, 01:04:01.860 --> 01:04:04.740 align:middle line:84% and everything this T bar dominates, so all of this. 01:04:04.740 --> 01:04:09.360 align:middle line:84% So "John" c-commands this position, "himself." 01:04:09.360 --> 01:04:10.560 align:middle line:90% Is this just reconstruction? 01:04:10.560 --> 01:04:11.490 align:middle line:90% Can "John" bind this? 01:04:11.490 --> 01:04:14.490 align:middle line:90% No, it's too far away, right? 01:04:14.490 --> 01:04:16.548 align:middle line:90% And so we can see that here. 01:04:16.548 --> 01:04:18.090 align:middle line:84% You can't say "John thought that Mary 01:04:18.090 --> 01:04:19.980 align:middle line:90% bought a picture of himself." 01:04:19.980 --> 01:04:22.260 align:middle line:84% So this position is too far away. 01:04:22.260 --> 01:04:25.140 align:middle line:84% And this position "John" doesn't c-command. 01:04:25.140 --> 01:04:29.650 align:middle line:84% If these are the only positions, then we are doomed. 01:04:29.650 --> 01:04:33.510 align:middle line:84% "Himself" can't be bound happily in either of those positions, 01:04:33.510 --> 01:04:34.530 align:middle line:90% doomed. 01:04:34.530 --> 01:04:38.130 align:middle line:84% But if "which picture of himself" 01:04:38.130 --> 01:04:44.910 align:middle line:84% spent some time here, well, then "John" c-commands "himself" 01:04:44.910 --> 01:04:47.790 align:middle line:84% when "which picture of himself" is here, right? 01:04:47.790 --> 01:04:51.360 align:middle line:84% This is inside this T bar, "John"'s sister, yeah? 01:04:51.360 --> 01:04:54.570 align:middle line:84% It's dominated by all the same nodes that dominate "John." 01:04:54.570 --> 01:05:00.540 align:middle line:84% And it's not contained in any TP that excludes "John." 01:05:00.540 --> 01:05:03.810 align:middle line:84% Remember, our conditions on how far away things can be 01:05:03.810 --> 01:05:05.963 align:middle line:84% and still bind say that anaphors need 01:05:05.963 --> 01:05:07.380 align:middle line:84% to have binders that are contained 01:05:07.380 --> 01:05:09.090 align:middle line:84% in all the TPs that contain them, 01:05:09.090 --> 01:05:12.190 align:middle line:90% that aren't excluded by any TPs. 01:05:12.190 --> 01:05:16.020 align:middle line:84% So if we do successive cyclic movement, which Dinka already 01:05:16.020 --> 01:05:17.940 align:middle line:84% told us that we should, if movement 01:05:17.940 --> 01:05:21.450 align:middle line:84% happens not in one big jump, but in two small jumps, 01:05:21.450 --> 01:05:24.090 align:middle line:84% then we have an account of this, the fact that the sentence is 01:05:24.090 --> 01:05:25.110 align:middle line:90% OK. 01:05:25.110 --> 01:05:29.070 align:middle line:84% The sentence is OK because "himself," yes, it's been here. 01:05:29.070 --> 01:05:32.160 align:middle line:84% Yes, it's up there, but it's also been here. 01:05:32.160 --> 01:05:36.450 align:middle line:84% And the fact that it spent time here is why the sentence is OK. 01:05:36.450 --> 01:05:37.950 align:middle line:84% Binding theory gives us a new reason 01:05:37.950 --> 01:05:42.460 align:middle line:84% to believe that, cool, so some of the cool things binding 01:05:42.460 --> 01:05:43.210 align:middle line:90% theory is for. 01:05:43.210 --> 01:05:43.710 align:middle line:90% Yeah. 01:05:43.710 --> 01:05:47.050 align:middle line:84% AUDIENCE: So it's like you go to multiple museums in one day, 01:05:47.050 --> 01:05:49.310 align:middle line:84% they give you a little stamp to get in the museum. 01:05:49.310 --> 01:05:49.750 align:middle line:90% NORVIN RICHARDS: Yes. 01:05:49.750 --> 01:05:51.145 align:middle line:84% AUDIENCE: And so at the end, you look a little bit different. 01:05:51.145 --> 01:05:53.800 align:middle line:84% You all have stamps to show that you've 01:05:53.800 --> 01:05:56.310 align:middle line:90% been to multiple museums. 01:05:56.310 --> 01:05:58.060 align:middle line:84% NORVIN RICHARDS: Yes, it's just like that. 01:05:58.060 --> 01:06:00.170 align:middle line:90% Yes. 01:06:00.170 --> 01:06:00.910 align:middle line:90% Yes. 01:06:00.910 --> 01:06:03.260 align:middle line:84% So "himself" has been here, and it's been up there. 01:06:03.260 --> 01:06:04.360 align:middle line:90% But it's also been here. 01:06:04.360 --> 01:06:06.640 align:middle line:84% And the fact that it was here meant that it got to go 01:06:06.640 --> 01:06:10.860 align:middle line:84% see the T-Rex skeleton while it was here. 01:06:10.860 --> 01:06:12.570 align:middle line:90% Yeah? 01:06:12.570 --> 01:06:13.200 align:middle line:90% OK, cool. 01:06:13.200 --> 01:06:16.020 align:middle line:90% 01:06:16.020 --> 01:06:18.120 align:middle line:84% So movement is successive-cyclic-- new reason 01:06:18.120 --> 01:06:18.900 align:middle line:90% to believe that. 01:06:18.900 --> 01:06:22.420 align:middle line:90% 01:06:22.420 --> 01:06:25.210 align:middle line:84% We have 15 minutes left, and I don't 01:06:25.210 --> 01:06:27.340 align:middle line:84% think I can do this next thing in 15 minutes. 01:06:27.340 --> 01:06:28.720 align:middle line:90% It involves guards. 01:06:28.720 --> 01:06:29.530 align:middle line:90% No, maybe I can. 01:06:29.530 --> 01:06:31.360 align:middle line:84% Let's start at least, and then this 01:06:31.360 --> 01:06:34.580 align:middle line:84% will give you something to think about until we get to Thursday. 01:06:34.580 --> 01:06:36.178 align:middle line:90% So consider this sentence. 01:06:36.178 --> 01:06:37.720 align:middle line:84% "Two guards seem to me to be standing 01:06:37.720 --> 01:06:39.880 align:middle line:90% in front of every building." 01:06:39.880 --> 01:06:42.310 align:middle line:90% Is that ambiguous? 01:06:42.310 --> 01:06:44.350 align:middle line:84% Yeah, it's ambiguous in a familiar way, right? 01:06:44.350 --> 01:06:48.870 align:middle line:84% It either means each building is guarded by two guards 01:06:48.870 --> 01:06:53.068 align:middle line:84% as far as I can tell, or there are two really weird guards 01:06:53.068 --> 01:06:54.860 align:middle line:84% who are guarding all the buildings at once. 01:06:54.860 --> 01:06:56.402 align:middle line:84% Yeah, it has those readings at least. 01:06:56.402 --> 01:06:59.280 align:middle line:90% 01:06:59.280 --> 01:07:00.330 align:middle line:90% How about this sentence? 01:07:00.330 --> 01:07:01.800 align:middle line:84% "Two guards seem to themselves"-- 01:07:01.800 --> 01:07:03.540 align:middle line:90% [LAUGHTER] 01:07:03.540 --> 01:07:05.400 align:middle line:90% Sorry. 01:07:05.400 --> 01:07:07.320 align:middle line:84% "Two guards seem to themselves to be standing 01:07:07.320 --> 01:07:08.487 align:middle line:90% in front of every building." 01:07:08.487 --> 01:07:11.550 align:middle line:90% 01:07:11.550 --> 01:07:14.070 align:middle line:90% This is not ambiguous. 01:07:14.070 --> 01:07:16.410 align:middle line:84% This means that there are two guards who 01:07:16.410 --> 01:07:19.320 align:middle line:90% need to be taken off duty. 01:07:19.320 --> 01:07:21.840 align:middle line:84% There are two guards who are having hallucinations 01:07:21.840 --> 01:07:25.007 align:middle line:90% of being very, very large. 01:07:25.007 --> 01:07:26.340 align:middle line:90% That's what this sentence means. 01:07:26.340 --> 01:07:28.798 align:middle line:84% It doesn't mean-- and there's something else it could mean. 01:07:28.798 --> 01:07:32.618 align:middle line:84% It could mean this building has two guards who 01:07:32.618 --> 01:07:34.410 align:middle line:84% think that they're standing in front of it. 01:07:34.410 --> 01:07:36.060 align:middle line:84% That building has two different guards 01:07:36.060 --> 01:07:38.018 align:middle line:84% who think that they're standing in front of it. 01:07:38.018 --> 01:07:39.813 align:middle line:84% That building has two other guards who 01:07:39.813 --> 01:07:40.980 align:middle line:90% are standing in front of it. 01:07:40.980 --> 01:07:42.720 align:middle line:84% In a sense, the reading that it actually has 01:07:42.720 --> 01:07:43.710 align:middle line:90% is the more reassuring one. 01:07:43.710 --> 01:07:45.793 align:middle line:84% It just means that there are two guards who really 01:07:45.793 --> 01:07:48.330 align:middle line:84% need a psych profile, whereas on the other reading 01:07:48.330 --> 01:07:50.850 align:middle line:84% there are however many buildings there are times two. 01:07:50.850 --> 01:07:54.040 align:middle line:84% There are that many guards who are in psychological trouble. 01:07:54.040 --> 01:07:56.880 align:middle line:84% So it doesn't mean all of the things that it could mean. 01:07:56.880 --> 01:08:00.990 align:middle line:84% And you might be able to guess the problem is themselves. 01:08:00.990 --> 01:08:04.200 align:middle line:84% That's what's going to prevent the other reading. 01:08:04.200 --> 01:08:06.720 align:middle line:84% But for reasons of time, let's let it prevent 01:08:06.720 --> 01:08:08.940 align:middle line:90% the other reading on Thursday. 01:08:08.940 --> 01:08:10.500 align:middle line:84% So come back on Thursday, and I'll 01:08:10.500 --> 01:08:13.563 align:middle line:84% finish this point, another thing that binding theory is for. 01:08:13.563 --> 01:08:15.480 align:middle line:84% We'll talk a little more about reconstruction, 01:08:15.480 --> 01:08:16.290 align:middle line:90% exactly what it is. 01:08:16.290 --> 01:08:17.335 align:middle line:90% Because it's fascinating. 01:08:17.335 --> 01:08:19.710 align:middle line:84% There's a lot of really cool work on trying to figure out 01:08:19.710 --> 01:08:20.989 align:middle line:90% what the heck it means. 01:08:20.989 --> 01:08:21.689 align:middle line:90% OK? 01:08:21.689 --> 01:08:24.060 align:middle line:90% Questions about anything else? 01:08:24.060 --> 01:08:25.050 align:middle line:90% All right, good. 01:08:25.050 --> 01:08:29.990 align:middle line:84% Go forth and think about linguistics. 01:08:29.990 --> 01:08:35.000 align:middle line:90%