WEBVTT 00:00:00.000 --> 00:00:10.010 align:middle line:84% [SQUEAKING] [RUSTLING] [CLICKING] 00:00:10.010 --> 00:00:11.870 align:middle line:84% NORVIN RICHARDS: Early on in this class, 00:00:11.870 --> 00:00:15.320 align:middle line:84% I think, I gave you examples-- it might have exactly 00:00:15.320 --> 00:00:16.309 align:middle line:90% been these examples. 00:00:16.309 --> 00:00:18.350 align:middle line:84% I meant to look back at the old slides to see. 00:00:18.350 --> 00:00:20.780 align:middle line:84% But it was something with this character anyway. 00:00:20.780 --> 00:00:22.370 align:middle line:90% I invited you to tell me-- 00:00:22.370 --> 00:00:27.110 align:middle line:84% if I ask you a question like "Who did Biden ask to see him?" 00:00:27.110 --> 00:00:29.180 align:middle line:90% I asked you, who's "him"? 00:00:29.180 --> 00:00:30.650 align:middle line:90% And I think, at the time-- 00:00:30.650 --> 00:00:33.410 align:middle line:84% who knows, we could do this exercise again and find out 00:00:33.410 --> 00:00:37.760 align:middle line:84% whether your minds have been warped by a semester of 24.900. 00:00:37.760 --> 00:00:39.410 align:middle line:84% I think, at the time, what you said 00:00:39.410 --> 00:00:41.600 align:middle line:84% was that "him" could refer to Biden. 00:00:41.600 --> 00:00:43.457 align:middle line:84% It was possible for this to mean "who"'s 00:00:43.457 --> 00:00:45.290 align:middle line:84% the person such that Biden asked that person 00:00:45.290 --> 00:00:48.710 align:middle line:84% to come to his, that is Biden's, office and see him, right? 00:00:48.710 --> 00:00:50.510 align:middle line:90% It could mean that. 00:00:50.510 --> 00:00:55.190 align:middle line:84% And then so it could also refer to someone else, of course. 00:00:55.190 --> 00:00:56.540 align:middle line:90% But it could refer to Biden. 00:00:56.540 --> 00:00:58.900 align:middle line:84% We weren't doing this at the time, 00:00:58.900 --> 00:01:00.150 align:middle line:90% at the beginning of the class. 00:01:00.150 --> 00:01:02.990 align:middle line:84% But we can represent that with these subscripts, 00:01:02.990 --> 00:01:05.750 align:middle line:84% since we've since talked about binding theory. 00:01:05.750 --> 00:01:09.475 align:middle line:84% This is the kind of fact we're in a position to talk about. 00:01:09.475 --> 00:01:11.600 align:middle line:84% And then what we also did at the beginning of class 00:01:11.600 --> 00:01:14.780 align:middle line:84% was to think also about questions like "Why did Biden 00:01:14.780 --> 00:01:16.370 align:middle line:90% ask to see him?" 00:01:16.370 --> 00:01:18.350 align:middle line:84% And what you all agreed at the time-- 00:01:18.350 --> 00:01:21.798 align:middle line:84% and feel free to jump in and stop me if you-- oh, 00:01:21.798 --> 00:01:22.840 align:middle line:90% Faith's about to stop me. 00:01:22.840 --> 00:01:23.750 align:middle line:84% AUDIENCE: I was just going to be like, I 00:01:23.750 --> 00:01:25.208 align:middle line:84% feel like now with everything we've 00:01:25.208 --> 00:01:27.440 align:middle line:84% learned about wh-movement, it feels wrong 00:01:27.440 --> 00:01:30.190 align:middle line:84% because you're moving something out of an island, I think. 00:01:30.190 --> 00:01:32.683 align:middle line:84% But you're keeping it there at the same time. 00:01:32.683 --> 00:01:34.400 align:middle line:84% Like "him," I feel like it was more 00:01:34.400 --> 00:01:37.750 align:middle line:84% appropriate to omit "him" because you're 00:01:37.750 --> 00:01:38.962 align:middle line:90% moving the "who" anyway. 00:01:38.962 --> 00:01:40.420 align:middle line:84% NORVIN RICHARDS: Oh, the first one? 00:01:40.420 --> 00:01:41.920 align:middle line:90% "Who did Biden ask to see him?" 00:01:41.920 --> 00:01:45.580 align:middle line:84% Oh, so in the first one, "Who did Biden ask to see him?" 00:01:45.580 --> 00:01:51.970 align:middle line:84% is supposed-- the answer is supposed to be the-- 00:01:51.970 --> 00:01:55.600 align:middle line:84% whatever, the head of the Joint Chiefs of Staff. 00:01:55.600 --> 00:01:57.910 align:middle line:84% He asked the head of the Joint Chiefs of Staff 00:01:57.910 --> 00:01:59.620 align:middle line:90% to come and see him, Biden. 00:01:59.620 --> 00:02:01.810 align:middle line:84% So Biden's worried about something 00:02:01.810 --> 00:02:03.790 align:middle line:84% that he wants that guy's advice on. 00:02:03.790 --> 00:02:06.280 align:middle line:84% And so he asked him to come see him. 00:02:06.280 --> 00:02:09.820 align:middle line:84% I think that's an OK question that you can ask that way. 00:02:09.820 --> 00:02:12.370 align:middle line:84% Yes, so you have been warped, you specifically, 00:02:12.370 --> 00:02:14.350 align:middle line:90% by a semester of 24.900. 00:02:14.350 --> 00:02:17.380 align:middle line:84% You're starting to fear islands everywhere, which is good. 00:02:17.380 --> 00:02:19.240 align:middle line:84% It's an important part of your training 00:02:19.240 --> 00:02:22.690 align:middle line:90% as a syntactician, yeah. 00:02:22.690 --> 00:02:23.200 align:middle line:90% Cool. 00:02:23.200 --> 00:02:25.812 align:middle line:84% So the first one then, "him" can refer to Biden, 00:02:25.812 --> 00:02:26.770 align:middle line:90% but it doesn't have to. 00:02:26.770 --> 00:02:28.395 align:middle line:84% For the second one, I think what people 00:02:28.395 --> 00:02:31.300 align:middle line:84% said at the time was that it wasn't possible for "him" 00:02:31.300 --> 00:02:33.190 align:middle line:90% to refer to Biden anymore. 00:02:33.190 --> 00:02:36.010 align:middle line:84% So "Why did Biden ask to see him?" 00:02:36.010 --> 00:02:37.540 align:middle line:84% That has to be, "Why did Biden ask 00:02:37.540 --> 00:02:40.120 align:middle line:84% to see some person, some particular person, 00:02:40.120 --> 00:02:41.440 align:middle line:90% but not Biden?" 00:02:41.440 --> 00:02:43.690 align:middle line:84% Do people still have that feeling? 00:02:43.690 --> 00:02:47.570 align:middle line:84% It's been a whole semester of suffering through this stuff. 00:02:47.570 --> 00:02:48.560 align:middle line:90% No? 00:02:48.560 --> 00:02:49.930 align:middle line:90% Yes, many of you do. 00:02:49.930 --> 00:02:51.680 align:middle line:84% Some of you, it's too early in the morning 00:02:51.680 --> 00:02:54.030 align:middle line:84% to have feelings about binding theory. 00:02:54.030 --> 00:02:55.790 align:middle line:84% You don't have to have feelings exactly. 00:02:55.790 --> 00:02:58.200 align:middle line:90% I'll settle for opinions. 00:02:58.200 --> 00:03:01.280 align:middle line:84% So I think there's this contrast between these sentences. 00:03:01.280 --> 00:03:03.380 align:middle line:84% And you know what I said at the beginning of class 00:03:03.380 --> 00:03:05.750 align:middle line:84% was look, these sentences, they look pretty darn 00:03:05.750 --> 00:03:09.410 align:middle line:84% similar-- they only vary in, like, one vowel, right? 00:03:09.410 --> 00:03:12.020 align:middle line:84% It's the difference between "who" and "why." 00:03:12.020 --> 00:03:14.395 align:middle line:84% And yet, you make that small change 00:03:14.395 --> 00:03:15.770 align:middle line:84% to the first word in the sentence 00:03:15.770 --> 00:03:18.230 align:middle line:84% and the last word in the sentence changes 00:03:18.230 --> 00:03:20.060 align:middle line:90% who it can refer to. 00:03:20.060 --> 00:03:21.900 align:middle line:84% Pretty complicated array of facts. 00:03:21.900 --> 00:03:24.307 align:middle line:84% And I think what I also promised, incautiously, 00:03:24.307 --> 00:03:26.390 align:middle line:84% was that by the end of class, you would understand 00:03:26.390 --> 00:03:28.200 align:middle line:90% why it worked this way. 00:03:28.200 --> 00:03:29.840 align:middle line:84% And so I just wanted to stop and talk 00:03:29.840 --> 00:03:32.060 align:middle line:84% about why it worked this way, partly 00:03:32.060 --> 00:03:33.560 align:middle line:84% as a way of segueing into what we're 00:03:33.560 --> 00:03:37.010 align:middle line:84% going to talk about today, which is language acquisition. 00:03:37.010 --> 00:03:40.600 align:middle line:90% So why does it work this way? 00:03:40.600 --> 00:03:42.505 align:middle line:84% So first, as Faith just said, one 00:03:42.505 --> 00:03:44.380 align:middle line:84% of the things we know about the first example 00:03:44.380 --> 00:03:47.860 align:middle line:84% is that "who" is moving, it's starting off as the object of 00:03:47.860 --> 00:03:48.550 align:middle line:90% "ask." 00:03:48.550 --> 00:03:52.480 align:middle line:84% So "Biden asked the chairman to see him." 00:03:52.480 --> 00:03:54.120 align:middle line:84% "The chairman" is the object of "ask." 00:03:54.120 --> 00:03:55.953 align:middle line:84% Instead of "the chairman," you've got "who." 00:03:55.953 --> 00:03:57.640 align:middle line:84% And "who" is undergoing wh-movement. 00:03:57.640 --> 00:04:00.670 align:middle line:84% "Why" is presumably also undergoing wh-movement. 00:04:00.670 --> 00:04:02.770 align:middle line:84% Let's not worry about where it's coming from. 00:04:02.770 --> 00:04:05.490 align:middle line:90% 00:04:05.490 --> 00:04:07.910 align:middle line:84% And the other thing that maybe you're 00:04:07.910 --> 00:04:11.030 align:middle line:84% in a position to believe by now-- 00:04:11.030 --> 00:04:13.940 align:middle line:84% we talked about this when we were doing binding theory-- 00:04:13.940 --> 00:04:16.250 align:middle line:84% was that these infinitival clauses, although they 00:04:16.250 --> 00:04:18.709 align:middle line:84% don't appear to have subjects, that they actually do 00:04:18.709 --> 00:04:19.459 align:middle line:90% have subjects. 00:04:19.459 --> 00:04:21.470 align:middle line:90% Their subjects are just null. 00:04:21.470 --> 00:04:22.670 align:middle line:90% We were calling it "PRO." 00:04:22.670 --> 00:04:26.270 align:middle line:84% It's this null pronoun that refers back, in both cases, 00:04:26.270 --> 00:04:28.230 align:middle line:84% to something in the higher clause. 00:04:28.230 --> 00:04:33.080 align:middle line:84% So in the first example, "Who did Biden ask to see him?" 00:04:33.080 --> 00:04:35.900 align:middle line:84% We consider the non-wh-moved version of this-- 00:04:35.900 --> 00:04:46.100 align:middle line:84% "Biden asked Harris PRO to see him." 00:04:46.100 --> 00:04:49.100 align:middle line:90% 00:04:49.100 --> 00:04:52.160 align:middle line:84% Here, PRO refers to Harris, right? 00:04:52.160 --> 00:04:55.550 align:middle line:84% This means Biden asked Harris to do the following thing. 00:04:55.550 --> 00:04:57.693 align:middle line:84% He wanted Harris to come see him. 00:04:57.693 --> 00:04:58.610 align:middle line:90% Does that sound right? 00:04:58.610 --> 00:04:59.580 align:middle line:90% That's what that means. 00:04:59.580 --> 00:05:02.630 align:middle line:84% So that PRO is an instance of what we called object control. 00:05:02.630 --> 00:05:05.420 align:middle line:84% PRO is referring back to an object in the higher clause. 00:05:05.420 --> 00:05:06.666 align:middle line:90% Yeah, Joseph? 00:05:06.666 --> 00:05:07.540 align:middle line:90% AUDIENCE: Sorry. 00:05:07.540 --> 00:05:08.582 align:middle line:90% NORVIN RICHARDS: It's OK. 00:05:08.582 --> 00:05:11.120 align:middle line:84% AUDIENCE: PRO in to satisfy the EPP. 00:05:11.120 --> 00:05:12.620 align:middle line:84% NORVIN RICHARDS: Yeah, that's right. 00:05:12.620 --> 00:05:13.610 align:middle line:90% That's what it was for. 00:05:13.610 --> 00:05:15.500 align:middle line:90% It satisfies the EPP. 00:05:15.500 --> 00:05:18.290 align:middle line:84% C needs a subject and doesn't appear to have one. 00:05:18.290 --> 00:05:21.200 align:middle line:84% And the claim back then was it really does have one. 00:05:21.200 --> 00:05:23.433 align:middle line:84% We went through this idea very quickly. 00:05:23.433 --> 00:05:25.100 align:middle line:84% One of the things we discovered about it 00:05:25.100 --> 00:05:29.000 align:middle line:84% was that it made the statement of binding theory simpler. 00:05:29.000 --> 00:05:31.670 align:middle line:84% It made it easier to understand why 00:05:31.670 --> 00:05:33.920 align:middle line:84% pronouns and anaphores inside infinitives 00:05:33.920 --> 00:05:35.480 align:middle line:90% act the way they do. 00:05:35.480 --> 00:05:37.890 align:middle line:84% And this is, in a way, an instance of that. 00:05:37.890 --> 00:05:40.130 align:middle line:90% So this PRO is referring back-- 00:05:40.130 --> 00:05:42.710 align:middle line:84% I'll give it the same subscript that it's got up there. 00:05:42.710 --> 00:05:46.355 align:middle line:84% If we change "Harris" to "who" and move it-- 00:05:46.355 --> 00:05:48.890 align:middle line:90% 00:05:48.890 --> 00:05:52.100 align:middle line:90% "Who did Biden ask to see him?" 00:05:52.100 --> 00:05:56.180 align:middle line:84% This PRO is controlled-- we say it refers back 00:05:56.180 --> 00:05:59.330 align:middle line:84% to the object of the higher clause. 00:05:59.330 --> 00:06:01.550 align:middle line:84% In the second example, the higher verb, "ask," 00:06:01.550 --> 00:06:02.575 align:middle line:90% doesn't have an object. 00:06:02.575 --> 00:06:04.680 align:middle line:84% It doesn't have to have an object. 00:06:04.680 --> 00:06:07.040 align:middle line:84% So you can say "Biden asks Harris to see him." 00:06:07.040 --> 00:06:15.230 align:middle line:84% Can also say "Biden asked--" this is with no object-- 00:06:15.230 --> 00:06:15.995 align:middle line:90% "to see him." 00:06:15.995 --> 00:06:18.870 align:middle line:90% 00:06:18.870 --> 00:06:23.280 align:middle line:84% And then the PRO gets controlled by the subject. 00:06:23.280 --> 00:06:25.278 align:middle line:84% So "ask," when it has an object, it's 00:06:25.278 --> 00:06:26.820 align:middle line:84% what's called an object control verb. 00:06:26.820 --> 00:06:28.828 align:middle line:84% The object of "ask" controls PRO. 00:06:28.828 --> 00:06:30.870 align:middle line:84% When it doesn't have an object, it's the subject. 00:06:30.870 --> 00:06:34.035 align:middle line:84% I think I said when we very briefly talked about this, 00:06:34.035 --> 00:06:38.670 align:middle line:84% it sure would be nice to understand why a verb-- 00:06:38.670 --> 00:06:42.390 align:middle line:84% what it is that conditions, which thing controls PRO. 00:06:42.390 --> 00:06:44.970 align:middle line:84% That's a topic of serious research. 00:06:44.970 --> 00:06:47.430 align:middle line:84% People try to figure out what the rules for that are. 00:06:47.430 --> 00:06:50.575 align:middle line:90% And it's not simple. 00:06:50.575 --> 00:06:52.450 align:middle line:84% So yeah, this is a way of representing what's 00:06:52.450 --> 00:06:53.770 align:middle line:90% going on in these sentences. 00:06:53.770 --> 00:06:57.160 align:middle line:84% Those embedded clauses have a subject, 00:06:57.160 --> 00:06:59.200 align:middle line:84% a subject you can't see, that refers back 00:06:59.200 --> 00:07:01.600 align:middle line:84% either to the object of "ask," in the first example, 00:07:01.600 --> 00:07:06.720 align:middle line:84% or to the subject of "ask," Biden, in the second example. 00:07:06.720 --> 00:07:08.860 align:middle line:90% This has been review, sort of. 00:07:08.860 --> 00:07:10.360 align:middle line:90% Does this make sense? 00:07:10.360 --> 00:07:12.540 align:middle line:90% Do people sort of remember this? 00:07:12.540 --> 00:07:14.570 align:middle line:90% No? 00:07:14.570 --> 00:07:16.400 align:middle line:84% And this, finally, allows us to understand 00:07:16.400 --> 00:07:18.920 align:middle line:84% why the pronoun at the end is behaving the way it is. 00:07:18.920 --> 00:07:22.370 align:middle line:84% So there's this general principle, principle B, 00:07:22.370 --> 00:07:25.490 align:middle line:84% that says pronouns have to be free in a certain domain. 00:07:25.490 --> 00:07:27.440 align:middle line:84% That is, they have to not have a binder, 00:07:27.440 --> 00:07:29.690 align:middle line:84% not have something that c-commands them and corefers 00:07:29.690 --> 00:07:30.320 align:middle line:90% with them. 00:07:30.320 --> 00:07:31.900 align:middle line:90% That's too close. 00:07:31.900 --> 00:07:34.790 align:middle line:84% We have to figure out what "too close" means. 00:07:34.790 --> 00:07:36.260 align:middle line:84% We can account for these data if we 00:07:36.260 --> 00:07:39.260 align:middle line:84% decide that the domain in which the pronoun has to be free 00:07:39.260 --> 00:07:41.360 align:middle line:84% is the domain that I've got brackets around, 00:07:41.360 --> 00:07:43.160 align:middle line:90% the embedded clause. 00:07:43.160 --> 00:07:46.470 align:middle line:84% So it has to be free in that clause. 00:07:46.470 --> 00:07:47.180 align:middle line:90% I don't care. 00:07:47.180 --> 00:07:49.230 align:middle line:90% Go away. 00:07:49.230 --> 00:07:50.240 align:middle line:90% No. 00:07:50.240 --> 00:07:53.370 align:middle line:84% It has to be free in that clause. 00:07:53.370 --> 00:07:54.280 align:middle line:90% What does that mean? 00:07:54.280 --> 00:07:58.050 align:middle line:84% Well, in the first example, it just 00:07:58.050 --> 00:08:00.998 align:middle line:84% means that "him" has to not refer to Harris. 00:08:00.998 --> 00:08:02.790 align:middle line:84% Well, there are lots of good reasons for it 00:08:02.790 --> 00:08:03.748 align:middle line:90% not to refer to Harris. 00:08:03.748 --> 00:08:06.195 align:middle line:90% I should have used a male name. 00:08:06.195 --> 00:08:09.150 align:middle line:84% It can't refer to the object of the higher clause 00:08:09.150 --> 00:08:12.030 align:middle line:84% because that object is controlling this pro. 00:08:12.030 --> 00:08:17.100 align:middle line:84% So if we kept Harris, if we said "Biden 00:08:17.100 --> 00:08:19.110 align:middle line:84% asked Harris to see him," then "him," of course, 00:08:19.110 --> 00:08:19.830 align:middle line:90% couldn't refer to Harris. 00:08:19.830 --> 00:08:22.140 align:middle line:84% But if you said "Biden asked Harris to see her," 00:08:22.140 --> 00:08:24.570 align:middle line:84% then "her" still couldn't refer to Harris-- 00:08:24.570 --> 00:08:27.450 align:middle line:84% background knowledge for anybody who has forgotten this Harris 00:08:27.450 --> 00:08:29.850 align:middle line:90% is female. 00:08:29.850 --> 00:08:31.350 align:middle line:84% "Her" still couldn't refer to Harris 00:08:31.350 --> 00:08:34.990 align:middle line:84% because "her" is too close to PRO would be the story, yeah. 00:08:34.990 --> 00:08:37.080 align:middle line:84% So that "him" is capable of referring to Biden 00:08:37.080 --> 00:08:40.747 align:middle line:84% because "Biden" is outside the bracketed domain. 00:08:40.747 --> 00:08:42.539 align:middle line:84% What about the "him" in the second example? 00:08:42.539 --> 00:08:44.414 align:middle line:84% Well, "Biden" is outside the bracketed domain 00:08:44.414 --> 00:08:45.370 align:middle line:90% in that example, too. 00:08:45.370 --> 00:08:49.350 align:middle line:84% But there is a PRO in the bracketed domain, 00:08:49.350 --> 00:08:51.870 align:middle line:84% the embedded clause, that is controlled by "Biden" 00:08:51.870 --> 00:08:52.920 align:middle line:90% and refers to Biden. 00:08:52.920 --> 00:08:55.352 align:middle line:84% So "him" cannot refer to Biden because there's-- not 00:08:55.352 --> 00:08:57.810 align:middle line:84% because "Biden" is too close, but because PRO is too close. 00:08:57.810 --> 00:09:00.510 align:middle line:84% PRO is inside the bracketed domain. 00:09:00.510 --> 00:09:01.990 align:middle line:90% And it binds him. 00:09:01.990 --> 00:09:03.580 align:middle line:90% It's too close. 00:09:03.580 --> 00:09:05.180 align:middle line:90% Yeah? 00:09:05.180 --> 00:09:06.500 align:middle line:90% People remember principle B? 00:09:06.500 --> 00:09:08.375 align:middle line:84% Principle B was meant to account for the fact 00:09:08.375 --> 00:09:22.400 align:middle line:84% that if I say "Biden thinks Harris likes him," 00:09:22.400 --> 00:09:25.520 align:middle line:90% that "him" can refer to Biden. 00:09:25.520 --> 00:09:27.950 align:middle line:90% "Biden" is far enough away. 00:09:27.950 --> 00:09:38.840 align:middle line:84% Whereas if I say "Harris thinks Biden likes him," now "him" 00:09:38.840 --> 00:09:39.860 align:middle line:90% cannot refer to Biden. 00:09:39.860 --> 00:09:41.210 align:middle line:90% "Biden" is too close. 00:09:41.210 --> 00:09:43.910 align:middle line:84% So we have a principle B that says "him"-- 00:09:43.910 --> 00:09:47.120 align:middle line:84% pronouns generally-- can't refer to a c-commanding noun 00:09:47.120 --> 00:09:48.530 align:middle line:90% phrase that's too close. 00:09:48.530 --> 00:09:49.910 align:middle line:90% We have to define "too close." 00:09:49.910 --> 00:09:53.370 align:middle line:84% But it's something like inside the same clause. 00:09:53.370 --> 00:09:55.640 align:middle line:84% So here, "Biden" is far enough away for "him" 00:09:55.640 --> 00:09:56.420 align:middle line:90% to refer to Biden. 00:09:56.420 --> 00:09:57.212 align:middle line:90% It doesn't have to. 00:09:57.212 --> 00:09:58.730 align:middle line:90% It could be somebody else. 00:09:58.730 --> 00:10:01.640 align:middle line:84% Here, "Biden" is too close and "him" can't refer to it. 00:10:01.640 --> 00:10:04.815 align:middle line:84% And something similar is going on in the slide. 00:10:04.815 --> 00:10:06.440 align:middle line:84% So this is me making good on my promise 00:10:06.440 --> 00:10:07.490 align:middle line:84% at the beginning of the semester. 00:10:07.490 --> 00:10:09.600 align:middle line:84% Beginning the semester, I said, by the end of the semester, 00:10:09.600 --> 00:10:10.940 align:middle line:90% you'll understand why this is. 00:10:10.940 --> 00:10:12.875 align:middle line:90% So now, you do, right? 00:10:12.875 --> 00:10:14.320 align:middle line:84% Do you have questions about this? 00:10:14.320 --> 00:10:15.195 align:middle line:90% Does this make sense? 00:10:15.195 --> 00:10:17.660 align:middle line:90% 00:10:17.660 --> 00:10:21.950 align:middle line:84% Well, so now we understand this-- yeah, yeah, 00:10:21.950 --> 00:10:23.690 align:middle line:90% I did all that. 00:10:23.690 --> 00:10:24.860 align:middle line:90% Now, we understand all this. 00:10:24.860 --> 00:10:30.600 align:middle line:84% But you know, all of you are exceptionally smart, right? 00:10:30.600 --> 00:10:33.450 align:middle line:84% And here you are getting an expensive university education, 00:10:33.450 --> 00:10:36.720 align:middle line:84% where someone is explaining this to you in great detail. 00:10:36.720 --> 00:10:39.760 align:middle line:84% And so that's why you understand what's going on here. 00:10:39.760 --> 00:10:43.492 align:middle line:84% But there's another sense in which you understood something 00:10:43.492 --> 00:10:44.700 align:middle line:90% about what was going on here. 00:10:44.700 --> 00:10:45.930 align:middle line:90% You had these intuitions. 00:10:45.930 --> 00:10:49.740 align:middle line:84% You knew what those pronouns could refer to before you ever 00:10:49.740 --> 00:10:51.210 align:middle line:90% set foot in this class. 00:10:51.210 --> 00:10:54.240 align:middle line:84% And people who don't have your advantages, people who are not 00:10:54.240 --> 00:10:56.340 align:middle line:84% getting expensive university educations, people 00:10:56.340 --> 00:10:59.760 align:middle line:84% who are not as smart as you, they also have those judgments. 00:10:59.760 --> 00:11:03.103 align:middle line:84% You can wander around Boston, ask any native speaker 00:11:03.103 --> 00:11:05.520 align:middle line:84% of English, they'll agree about these sentences, about who 00:11:05.520 --> 00:11:08.080 align:middle line:90% "him" refers to, yeah? 00:11:08.080 --> 00:11:13.140 align:middle line:84% So question-- how did that happen? 00:11:13.140 --> 00:11:17.070 align:middle line:84% How do you end up with this, in the sense, exceptionally 00:11:17.070 --> 00:11:18.960 align:middle line:90% complicated knowledge? 00:11:18.960 --> 00:11:21.540 align:middle line:90% How are you doing all of this? 00:11:21.540 --> 00:11:23.550 align:middle line:84% This is a subpart of what Chomsky 00:11:23.550 --> 00:11:26.190 align:middle line:90% likes to call Plato's problem. 00:11:26.190 --> 00:11:30.000 align:middle line:84% Plato, I guess, in his writings somewhere asked this question-- 00:11:30.000 --> 00:11:31.890 align:middle line:84% how is it that-- he wasn't talking 00:11:31.890 --> 00:11:33.210 align:middle line:90% about language specifically. 00:11:33.210 --> 00:11:35.830 align:middle line:84% How is it that we know so much about the world, 00:11:35.830 --> 00:11:39.180 align:middle line:84% despite the fact that our evidence is 00:11:39.180 --> 00:11:41.100 align:middle line:90% kind of random and spotty? 00:11:41.100 --> 00:11:44.040 align:middle line:84% It's not as if your parents ever sat you down and explained, 00:11:44.040 --> 00:11:48.360 align:middle line:84% "Oh by the way, if I say 'Biden asked to see him,' then 00:11:48.360 --> 00:11:50.843 align:middle line:90% 'him' can't refer to Biden." 00:11:50.843 --> 00:11:52.260 align:middle line:84% Nobody ever explained that to you. 00:11:52.260 --> 00:11:55.320 align:middle line:84% You just kind of figured it out, despite the lack 00:11:55.320 --> 00:11:58.110 align:middle line:90% of any relevant instruction-- 00:11:58.110 --> 00:12:01.260 align:middle line:84% not just this fact, but a whole bunch of complicated facts, all 00:12:01.260 --> 00:12:04.770 align:middle line:84% the complicated stuff we've been talking about in this class. 00:12:04.770 --> 00:12:07.830 align:middle line:84% Your parents never explained that to you, I assume. 00:12:07.830 --> 00:12:08.410 align:middle line:90% Who knows? 00:12:08.410 --> 00:12:09.910 align:middle line:84% Maybe there's somebody in this class 00:12:09.910 --> 00:12:12.360 align:middle line:84% whose parents were exceptionally helpful about this stuff. 00:12:12.360 --> 00:12:15.520 align:middle line:84% But presumably, not all of you had your parents explain this 00:12:15.520 --> 00:12:16.020 align:middle line:90% to you. 00:12:16.020 --> 00:12:19.200 align:middle line:90% 00:12:19.200 --> 00:12:21.720 align:middle line:84% People who have studied the interactions of parents 00:12:21.720 --> 00:12:26.910 align:middle line:84% with children, it's pretty clear that parents very occasionally 00:12:26.910 --> 00:12:33.000 align:middle line:84% give children direct instruction about how to speak. 00:12:33.000 --> 00:12:35.170 align:middle line:84% But they usually have other kinds of priorities. 00:12:35.170 --> 00:12:38.490 align:middle line:84% So this is an actual dialogue between a child and the mother. 00:12:38.490 --> 00:12:41.100 align:middle line:84% The child says, "Mama isn't a boy, he a girl." 00:12:41.100 --> 00:12:42.945 align:middle line:90% The mother says, "That's right." 00:12:42.945 --> 00:12:45.300 align:middle line:84% The child says, and "Walt Disney comes on Tuesday." 00:12:45.300 --> 00:12:46.830 align:middle line:84% The mother says, "No, he does not." 00:12:46.830 --> 00:12:49.558 align:middle line:84% So this is an actual dialogue between a small child 00:12:49.558 --> 00:12:50.100 align:middle line:90% and a mother. 00:12:50.100 --> 00:12:51.683 align:middle line:84% And something that you notice about it 00:12:51.683 --> 00:12:54.300 align:middle line:84% is that if the child is relying on the parent 00:12:54.300 --> 00:12:58.560 align:middle line:84% to get explicit instruction about grammar, 00:12:58.560 --> 00:13:01.530 align:middle line:84% then the child is out of luck, right? 00:13:01.530 --> 00:13:06.090 align:middle line:84% So the child's first sentence has some grammatical problems. 00:13:06.090 --> 00:13:08.250 align:middle line:84% The child is using "he" to refer to-- 00:13:08.250 --> 00:13:10.410 align:middle line:90% to refer to the mother. 00:13:10.410 --> 00:13:12.235 align:middle line:84% The response that the mother gets-- 00:13:12.235 --> 00:13:13.860 align:middle line:84% the child gets from the mother for that 00:13:13.860 --> 00:13:16.050 align:middle line:90% is "That's right," right? 00:13:16.050 --> 00:13:19.170 align:middle line:84% Because the mother is not thinking too much 00:13:19.170 --> 00:13:21.570 align:middle line:90% about the grammar. 00:13:21.570 --> 00:13:24.030 align:middle line:84% The child has made an accurate observation about the world, 00:13:24.030 --> 00:13:25.250 align:middle line:90% right? 00:13:25.250 --> 00:13:27.177 align:middle line:84% We can worry later about pronouns. 00:13:27.177 --> 00:13:29.010 align:middle line:84% The point is the child is figuring something 00:13:29.010 --> 00:13:31.320 align:middle line:84% out about the gender of their mother. 00:13:31.320 --> 00:13:33.840 align:middle line:84% And then the child says this perfectly grammatical sentence 00:13:33.840 --> 00:13:34.260 align:middle line:90% at the end-- 00:13:34.260 --> 00:13:35.718 align:middle line:84% "And Walt Disney comes on Tuesday." 00:13:35.718 --> 00:13:38.010 align:middle line:84% And the mother says, "No, you're wrong," right? 00:13:38.010 --> 00:13:41.658 align:middle line:84% Well, OK, so apparently, the mother 00:13:41.658 --> 00:13:43.950 align:middle line:84% is not paying too much attention to stuff like grammar. 00:13:43.950 --> 00:13:47.250 align:middle line:84% The mother is paying attention to accuracy about the world. 00:13:47.250 --> 00:13:49.560 align:middle line:84% And this child is therefore going 00:13:49.560 --> 00:13:52.200 align:middle line:84% to grow up to be a normal and well-adjusted human being 00:13:52.200 --> 00:13:55.200 align:middle line:84% because that's what you want from your parents. 00:13:55.200 --> 00:13:59.190 align:middle line:84% It would be odd for the mother to say "No,, no, you're using 00:13:59.190 --> 00:14:00.810 align:middle line:90% the masculine pronoun for me. 00:14:00.810 --> 00:14:03.460 align:middle line:90% And that's not correct." 00:14:03.460 --> 00:14:05.910 align:middle line:84% There actually are dialogues-- if you've ever 00:14:05.910 --> 00:14:12.300 align:middle line:84% been around a small child, you know when parents do attempt 00:14:12.300 --> 00:14:18.420 align:middle line:84% to instruct their children in how to speak, it often-- it's 00:14:18.420 --> 00:14:21.420 align:middle line:84% not clear that the children are paying any attention. 00:14:21.420 --> 00:14:24.600 align:middle line:84% So here-- again, I swear I am not making this up. 00:14:24.600 --> 00:14:25.830 align:middle line:90% Somebody recorded this. 00:14:25.830 --> 00:14:28.290 align:middle line:84% A child said, "Nobody don't like me." 00:14:28.290 --> 00:14:32.130 align:middle line:84% And the father, who should really have had his fatherhood 00:14:32.130 --> 00:14:38.190 align:middle line:84% license taken away, said, "No, say 'Nobody likes me.'" 00:14:38.190 --> 00:14:41.070 align:middle line:84% This dialogue repeats six times on the tape. 00:14:41.070 --> 00:14:43.350 align:middle line:84% And then finally, the child says, 00:14:43.350 --> 00:14:46.080 align:middle line:90% "Oh, nobody don't likes me." 00:14:46.080 --> 00:14:47.730 align:middle line:90% And curtain. 00:14:47.730 --> 00:14:51.000 align:middle line:84% I don't know what happened after that. 00:14:51.000 --> 00:14:53.220 align:middle line:84% I mean, this particular dialogue is hopefully 00:14:53.220 --> 00:14:56.440 align:middle line:84% not one you'll often hear, but you do. 00:14:56.440 --> 00:14:59.940 align:middle line:84% There's another classic that somebody recorded. 00:14:59.940 --> 00:15:04.020 align:middle line:84% A child said, "Give me other one spoon. 00:15:04.020 --> 00:15:05.760 align:middle line:90% I want other one spoon," right? 00:15:05.760 --> 00:15:08.760 align:middle line:84% And the parent tries many times to get 00:15:08.760 --> 00:15:10.760 align:middle line:90% them to say "the other spoon." 00:15:10.760 --> 00:15:12.938 align:middle line:84% "No, say 'the other spoon.'" "Other one spoon." 00:15:12.938 --> 00:15:14.730 align:middle line:84% "Say 'the other spoon.'" "Other one spoon." 00:15:14.730 --> 00:15:16.515 align:middle line:84% "Say 'the other'"-- "No, other one spoon." 00:15:16.515 --> 00:15:17.890 align:middle line:84% They go back and forth like this. 00:15:17.890 --> 00:15:21.080 align:middle line:84% And the parent says one word at a time-- "the, the, 00:15:21.080 --> 00:15:23.880 align:middle line:90% other, other, spoon, spoon." 00:15:23.880 --> 00:15:25.860 align:middle line:84% "Yeah, the other spoon," says the child. 00:15:25.860 --> 00:15:29.150 align:middle line:84% "Now give me other one spoon?" asks the child. 00:15:29.150 --> 00:15:30.950 align:middle line:90% So parents do-- 00:15:30.950 --> 00:15:32.120 align:middle line:90% I have been a parent. 00:15:32.120 --> 00:15:33.830 align:middle line:84% If you've ever been around children-- so 00:15:33.830 --> 00:15:36.140 align:middle line:84% I can attest that this is how it works. 00:15:36.140 --> 00:15:40.080 align:middle line:84% Even if you are trying to tell children how to speak, 00:15:40.080 --> 00:15:42.320 align:middle line:90% they're not listening to you. 00:15:42.320 --> 00:15:44.720 align:middle line:90% And most of the time-- 00:15:44.720 --> 00:15:46.580 align:middle line:84% the first dialogue illustrates this-- 00:15:46.580 --> 00:15:48.753 align:middle line:84% the parents quite rightly are not really 00:15:48.753 --> 00:15:50.420 align:middle line:84% paying too much attention to the grammar 00:15:50.420 --> 00:15:51.712 align:middle line:90% of their children's utterances. 00:15:51.712 --> 00:15:53.930 align:middle line:84% They're just excited by the fact that their children 00:15:53.930 --> 00:15:55.105 align:middle line:90% are talking. 00:15:55.105 --> 00:15:58.620 align:middle line:90% 00:15:58.620 --> 00:16:02.760 align:middle line:84% So how are we learning all this complicated stuff? 00:16:02.760 --> 00:16:04.920 align:middle line:84% It's not because your parents teach it to you. 00:16:04.920 --> 00:16:07.620 align:middle line:84% Even when they try, you don't pay attention, apparently, 00:16:07.620 --> 00:16:09.990 align:middle line:90% at the relevant stage. 00:16:09.990 --> 00:16:13.615 align:middle line:84% And there's an answer that I've been pushing in this class. 00:16:13.615 --> 00:16:15.600 align:middle line:84% It's this idea of universal grammar, which 00:16:15.600 --> 00:16:18.540 align:middle line:84% is that human beings can't help building 00:16:18.540 --> 00:16:20.290 align:middle line:90% our languages in certain ways. 00:16:20.290 --> 00:16:24.268 align:middle line:84% And so you, of course, need some data 00:16:24.268 --> 00:16:26.310 align:middle line:84% in order to build your language because there are 00:16:26.310 --> 00:16:28.600 align:middle line:90% 6,000 some languages out there. 00:16:28.600 --> 00:16:30.690 align:middle line:84% So you need to learn which of those languages 00:16:30.690 --> 00:16:32.640 align:middle line:90% it is that you're learning. 00:16:32.640 --> 00:16:37.440 align:middle line:84% But languages come with certain properties sort 00:16:37.440 --> 00:16:39.900 align:middle line:90% of factory preset. 00:16:39.900 --> 00:16:42.600 align:middle line:84% You don't actually have to learn all this stuff 00:16:42.600 --> 00:16:45.900 align:middle line:84% about controlled PRO, and condition B, and so on. 00:16:45.900 --> 00:16:48.430 align:middle line:84% Universal languages just work that way. 00:16:48.430 --> 00:16:50.190 align:middle line:84% It's just a matter of learning which 00:16:50.190 --> 00:16:52.320 align:middle line:84% of the words in your language means "ask," 00:16:52.320 --> 00:16:54.570 align:middle line:84% and which one means "see," and which one means "him," 00:16:54.570 --> 00:16:55.123 align:middle line:90% and so on. 00:16:55.123 --> 00:16:56.790 align:middle line:84% And once you know those things, then you 00:16:56.790 --> 00:16:58.770 align:middle line:84% know that your language will work this way 00:16:58.770 --> 00:16:59.940 align:middle line:90% because you can't help it. 00:16:59.940 --> 00:17:03.520 align:middle line:90% That's the way languages work. 00:17:03.520 --> 00:17:06.960 align:middle line:84% So we've spent the whole semester 00:17:06.960 --> 00:17:09.460 align:middle line:84% trying to figure out exactly what it is that you know when 00:17:09.460 --> 00:17:10.710 align:middle line:90% you know your native language. 00:17:10.710 --> 00:17:14.013 align:middle line:84% And I hope I've convinced you that it's more complicated 00:17:14.013 --> 00:17:15.180 align:middle line:90% than you might have thought. 00:17:15.180 --> 00:17:18.000 align:middle line:84% You know quite a lot, lots of complicated stuff 00:17:18.000 --> 00:17:19.170 align:middle line:90% going on in your head. 00:17:19.170 --> 00:17:22.210 align:middle line:90% 00:17:22.210 --> 00:17:25.270 align:middle line:90% And so a question arises. 00:17:25.270 --> 00:17:26.859 align:middle line:90% What's innate? 00:17:26.859 --> 00:17:28.960 align:middle line:84% What is this stuff that's preset? 00:17:28.960 --> 00:17:30.280 align:middle line:90% And what is learned? 00:17:30.280 --> 00:17:32.020 align:middle line:84% And as I've just said, it's obvious 00:17:32.020 --> 00:17:34.390 align:middle line:84% that not everything is innate because there's 00:17:34.390 --> 00:17:36.280 align:middle line:84% more than one language out there. 00:17:36.280 --> 00:17:38.680 align:middle line:84% You've got to learn which string of sounds 00:17:38.680 --> 00:17:40.750 align:middle line:84% means "ask," which string of sounds 00:17:40.750 --> 00:17:43.340 align:middle line:84% means "cat," which string of sounds means "spoon," 00:17:43.340 --> 00:17:45.740 align:middle line:90% and so on. 00:17:45.740 --> 00:17:50.320 align:middle line:84% There's some evidence also that evidence matters, 00:17:50.320 --> 00:17:52.660 align:middle line:84% which comes from the idea of what's 00:17:52.660 --> 00:17:54.610 align:middle line:90% called a critical period. 00:17:54.610 --> 00:17:56.950 align:middle line:90% It looks as though-- 00:17:56.950 --> 00:17:59.590 align:middle line:84% it's very hard to get direct evidence about this, 00:17:59.590 --> 00:18:01.570 align:middle line:84% but it looks as though there may be 00:18:01.570 --> 00:18:04.000 align:middle line:84% a period in the life of a child when 00:18:04.000 --> 00:18:08.710 align:middle line:84% they're especially well-designed to acquire language. 00:18:08.710 --> 00:18:13.960 align:middle line:84% And after that period is over, they get worse at it. 00:18:13.960 --> 00:18:16.120 align:middle line:84% There's a lot of anecdotal evidence for this. 00:18:16.120 --> 00:18:20.907 align:middle line:84% All of us fluently learn one or more languages as a child. 00:18:20.907 --> 00:18:22.990 align:middle line:84% When you're a small child, you learn the languages 00:18:22.990 --> 00:18:24.580 align:middle line:90% that are around you. 00:18:24.580 --> 00:18:25.820 align:middle line:90% For me, it was one language. 00:18:25.820 --> 00:18:28.195 align:middle line:84% There are lots of people who learn more than one language 00:18:28.195 --> 00:18:30.190 align:middle line:90% growing up. 00:18:30.190 --> 00:18:34.390 align:middle line:84% But anecdotally, for most of us, once you're an adult, 00:18:34.390 --> 00:18:37.570 align:middle line:84% it's much harder to fluently pick up another language 00:18:37.570 --> 00:18:39.550 align:middle line:90% just by being exposed to it. 00:18:39.550 --> 00:18:42.520 align:middle line:84% Even if you imagine being surrounded 00:18:42.520 --> 00:18:44.680 align:middle line:84% by people who are willing to just talk to you all 00:18:44.680 --> 00:18:47.200 align:middle line:84% the time in another language, it's not clear 00:18:47.200 --> 00:18:50.620 align:middle line:84% that you would learn it in a matter of six years or seven 00:18:50.620 --> 00:18:53.770 align:middle line:84% years, which is what you did with your first language. 00:18:53.770 --> 00:18:58.760 align:middle line:84% There are some cases that bear directly on this hypothesis. 00:18:58.760 --> 00:19:01.420 align:middle line:84% They're sort of awful cases, but I feel the need 00:19:01.420 --> 00:19:03.080 align:middle line:90% to tell you about them. 00:19:03.080 --> 00:19:05.920 align:middle line:84% One of them is the story of a child who 00:19:05.920 --> 00:19:08.110 align:middle line:90% was given the pseudonym Genie. 00:19:08.110 --> 00:19:14.830 align:middle line:84% Genie was discovered, I think, in LA in 1970. 00:19:14.830 --> 00:19:20.620 align:middle line:84% She was the child of a mother who was almost blind. 00:19:20.620 --> 00:19:24.260 align:middle line:84% Her mother had a degenerative disease of her eyes. 00:19:24.260 --> 00:19:25.918 align:middle line:90% She was gradually going blind. 00:19:25.918 --> 00:19:27.460 align:middle line:84% And by the time Genie was discovered, 00:19:27.460 --> 00:19:30.070 align:middle line:90% she was almost completely blind. 00:19:30.070 --> 00:19:35.050 align:middle line:84% And her father was crazy, so probably 00:19:35.050 --> 00:19:36.670 align:middle line:90% paranoid schizophrenic. 00:19:36.670 --> 00:19:41.080 align:middle line:90% He pathologically hated noise. 00:19:41.080 --> 00:19:42.320 align:middle line:90% He couldn't stand noise. 00:19:42.320 --> 00:19:44.920 align:middle line:84% And so if you've ever been around small children, 00:19:44.920 --> 00:19:47.320 align:middle line:84% you know that they sometimes make noise. 00:19:47.320 --> 00:19:48.102 align:middle line:90% So he was-- 00:19:48.102 --> 00:19:50.560 align:middle line:84% I mean, talk about people who should have had their license 00:19:50.560 --> 00:19:52.480 align:middle line:90% to be a father taken away. 00:19:52.480 --> 00:19:58.420 align:middle line:84% Genie had one older sibling who died of exposure because they 00:19:58.420 --> 00:20:04.750 align:middle line:84% were kept in the garage of the house because they were noisy 00:20:04.750 --> 00:20:07.810 align:middle line:90% and the father couldn't take it. 00:20:07.810 --> 00:20:10.270 align:middle line:84% She had another older brother who did survive, 00:20:10.270 --> 00:20:11.140 align:middle line:90% I don't know how. 00:20:11.140 --> 00:20:13.960 align:middle line:84% Maybe her father wasn't quite as crazy when 00:20:13.960 --> 00:20:15.880 align:middle line:90% her older brother was born. 00:20:15.880 --> 00:20:17.992 align:middle line:84% But when Genie was born, they had 00:20:17.992 --> 00:20:20.200 align:middle line:84% learned not to keep the babies in the garage anymore. 00:20:20.200 --> 00:20:24.340 align:middle line:84% What they did was to keep her in a dark room strapped to a chair 00:20:24.340 --> 00:20:26.410 align:middle line:84% and to punish her for making sounds. 00:20:26.410 --> 00:20:28.660 align:middle line:90% So no one ever spoke to her. 00:20:28.660 --> 00:20:30.820 align:middle line:84% She apparently could sometimes hear piano music 00:20:30.820 --> 00:20:32.180 align:middle line:90% from a nearby house. 00:20:32.180 --> 00:20:37.040 align:middle line:84% So when she grew up, she always loved piano music. 00:20:37.040 --> 00:20:39.040 align:middle line:84% Her father and her older brother would sometimes 00:20:39.040 --> 00:20:40.510 align:middle line:90% bark at her like dogs. 00:20:40.510 --> 00:20:43.300 align:middle line:84% So she was always afraid of dogs. 00:20:43.300 --> 00:20:46.600 align:middle line:84% That went on for the first 13 years 00:20:46.600 --> 00:20:50.170 align:middle line:84% of her life, after which she was discovered and taken away 00:20:50.170 --> 00:20:53.630 align:middle line:90% from her parents. 00:20:53.630 --> 00:20:57.070 align:middle line:84% She was discovered because her mother somehow-- 00:20:57.070 --> 00:20:59.330 align:middle line:84% who as I said, was almost blind-- 00:20:59.330 --> 00:21:01.660 align:middle line:84% had gotten out of the house with Genie 00:21:01.660 --> 00:21:04.350 align:middle line:84% and was trying to find a government office where 00:21:04.350 --> 00:21:06.850 align:middle line:84% somebody could help her-- not with Genie, but with something 00:21:06.850 --> 00:21:08.267 align:middle line:84% having to do with their blindness. 00:21:08.267 --> 00:21:11.800 align:middle line:84% She was looking for an office that would help her with that 00:21:11.800 --> 00:21:13.690 align:middle line:84% and wandered by accident into an office that 00:21:13.690 --> 00:21:14.830 align:middle line:90% was about child welfare. 00:21:14.830 --> 00:21:16.630 align:middle line:84% And they immediately spotted that there 00:21:16.630 --> 00:21:19.480 align:middle line:90% was something off about Genie. 00:21:19.480 --> 00:21:23.620 align:middle line:84% So Genie was taken away from her parents-- 00:21:23.620 --> 00:21:29.260 align:middle line:84% sorry-- and put in a series of institutions, 00:21:29.260 --> 00:21:32.590 align:middle line:84% where lots of people did their best 00:21:32.590 --> 00:21:36.490 align:middle line:84% to help her recover from the first 13 years of her life. 00:21:36.490 --> 00:21:39.010 align:middle line:84% And she did, in a sense, recover, in the sense 00:21:39.010 --> 00:21:46.870 align:middle line:84% that she acquired the ability to stop attacking people when 00:21:46.870 --> 00:21:47.440 align:middle line:90% she saw them. 00:21:47.440 --> 00:21:52.840 align:middle line:84% And she was wild when she was first out of the house. 00:21:52.840 --> 00:21:54.970 align:middle line:84% And she learned to say some words. 00:21:54.970 --> 00:21:57.610 align:middle line:84% But she never learned to speak, in the sense 00:21:57.610 --> 00:21:59.740 align:middle line:84% of putting words together in anything 00:21:59.740 --> 00:22:00.760 align:middle line:90% that resembled grammar. 00:22:00.760 --> 00:22:04.040 align:middle line:90% So she was just saying words. 00:22:04.040 --> 00:22:06.490 align:middle line:84% So she knew words for things that she wanted. 00:22:06.490 --> 00:22:09.880 align:middle line:84% And insofar as it was possible to do intelligence tests 00:22:09.880 --> 00:22:12.790 align:middle line:84% on somebody who was as profoundly traumatized as she 00:22:12.790 --> 00:22:16.760 align:middle line:84% was, she scored as a perfectly intelligent person. 00:22:16.760 --> 00:22:19.898 align:middle line:84% She had a normal mind in all kinds of ways. 00:22:19.898 --> 00:22:21.190 align:middle line:90% But she never learned to speak. 00:22:21.190 --> 00:22:23.950 align:middle line:90% 00:22:23.950 --> 00:22:26.350 align:middle line:84% So that's one case that people say, ah, 00:22:26.350 --> 00:22:29.878 align:middle line:84% so if you deprive a child of linguistic information 00:22:29.878 --> 00:22:31.420 align:middle line:84% for the beginning of their life, then 00:22:31.420 --> 00:22:33.580 align:middle line:84% this has profound consequences for their ability 00:22:33.580 --> 00:22:35.770 align:middle line:90% to learn to speak. 00:22:35.770 --> 00:22:39.340 align:middle line:84% And it could be that that's what this shows. 00:22:39.340 --> 00:22:43.690 align:middle line:84% It could also be that if you terribly 00:22:43.690 --> 00:22:46.690 align:middle line:84% traumatize someone for the first 13 years of your life, 00:22:46.690 --> 00:22:48.970 align:middle line:84% that you damage their ability to learn to speak. 00:22:48.970 --> 00:22:52.540 align:middle line:84% As an example of the critical period, 00:22:52.540 --> 00:22:56.800 align:middle line:84% it lacks a certain scientific rigor, it seems to me. 00:22:56.800 --> 00:22:58.550 align:middle line:84% So I apologize for telling you this story. 00:22:58.550 --> 00:23:00.700 align:middle line:84% I feel as though anyone who takes intro linguistics 00:23:00.700 --> 00:23:04.125 align:middle line:90% should learn about Genie. 00:23:04.125 --> 00:23:05.750 align:middle line:84% Though, as I say, Genie is a pseudonym. 00:23:05.750 --> 00:23:09.390 align:middle line:90% I have no idea what her name is. 00:23:09.390 --> 00:23:10.980 align:middle line:84% There are better stories than that. 00:23:10.980 --> 00:23:14.190 align:middle line:90% Chelsea, also a pseudonym-- 00:23:14.190 --> 00:23:17.790 align:middle line:84% and actually, Chelsea's case is a case that used to be common-- 00:23:17.790 --> 00:23:20.620 align:middle line:84% not common-- used to be more common than it is now. 00:23:20.620 --> 00:23:22.158 align:middle line:84% There are safeguards in place now 00:23:22.158 --> 00:23:23.700 align:middle line:84% to stop this from happening as often. 00:23:23.700 --> 00:23:28.020 align:middle line:84% Chelsea is another girl who grew up in Northern California-- 00:23:28.020 --> 00:23:31.950 align:middle line:84% I don't know what it is about Northern California-- 00:23:31.950 --> 00:23:38.520 align:middle line:84% whose parents believed that she was profoundly developmentally 00:23:38.520 --> 00:23:41.050 align:middle line:90% disabled. 00:23:41.050 --> 00:23:43.050 align:middle line:84% This is not a story like the Genie story. 00:23:43.050 --> 00:23:45.570 align:middle line:84% So her parents were perfectly loving, caring people. 00:23:45.570 --> 00:23:48.420 align:middle line:84% They were doing their best to raise her. 00:23:48.420 --> 00:23:51.150 align:middle line:90% They treated her very well. 00:23:51.150 --> 00:23:53.700 align:middle line:90% She had a happy life. 00:23:53.700 --> 00:23:57.900 align:middle line:84% But she wasn't-- she was developmentally behind on all 00:23:57.900 --> 00:23:58.830 align:middle line:90% kinds of milestones. 00:23:58.830 --> 00:24:00.880 align:middle line:84% In particular, she wasn't learning to speak. 00:24:00.880 --> 00:24:06.810 align:middle line:84% And so her parents just did their best to deal with that. 00:24:06.810 --> 00:24:07.860 align:middle line:90% At some point-- 00:24:07.860 --> 00:24:10.440 align:middle line:84% I want to say when she was 12 or 13-- 00:24:10.440 --> 00:24:12.840 align:middle line:84% it was discovered that she wasn't developmentally 00:24:12.840 --> 00:24:13.590 align:middle line:90% disabled at all. 00:24:13.590 --> 00:24:17.490 align:middle line:84% She was deaf, that she was profoundly deaf. 00:24:17.490 --> 00:24:20.340 align:middle line:90% And her parents-- 00:24:20.340 --> 00:24:22.580 align:middle line:84% I mean, again, perfectly kind, loving, 00:24:22.580 --> 00:24:26.628 align:middle line:84% sort of unobservant people had never noticed this, I guess. 00:24:26.628 --> 00:24:28.170 align:middle line:84% So the fact that she wasn't speaking, 00:24:28.170 --> 00:24:31.050 align:middle line:84% it wasn't because she was developmentally disabled. 00:24:31.050 --> 00:24:34.590 align:middle line:84% It was because she hadn't heard any speech. 00:24:34.590 --> 00:24:36.393 align:middle line:84% And nobody had been signing to her 00:24:36.393 --> 00:24:38.310 align:middle line:84% because they didn't realize that she was deaf. 00:24:38.310 --> 00:24:41.410 align:middle line:84% So she hadn't been exposed to language in any form either. 00:24:41.410 --> 00:24:44.880 align:middle line:84% And as I say, it's a happier story than the Genie story 00:24:44.880 --> 00:24:47.988 align:middle line:84% because there was no actual abuse involved. 00:24:47.988 --> 00:24:49.530 align:middle line:84% And as I also said, this is something 00:24:49.530 --> 00:24:51.238 align:middle line:84% that happens much less common these days. 00:24:51.238 --> 00:24:53.820 align:middle line:84% So all of you hopefully were screened 00:24:53.820 --> 00:24:55.140 align:middle line:90% for deafness in schools. 00:24:55.140 --> 00:24:57.270 align:middle line:84% And this is one of the reasons that this happens, 00:24:57.270 --> 00:24:59.550 align:middle line:84% is to stop this kind of thing from happening, 00:24:59.550 --> 00:25:03.960 align:middle line:84% stop people from slipping through the cracks in this way. 00:25:03.960 --> 00:25:07.890 align:middle line:84% Chelsea, like Genie, wasn't exposed 00:25:07.890 --> 00:25:11.400 align:middle line:84% to language for the first 12 years or so of her life. 00:25:11.400 --> 00:25:14.370 align:middle line:84% And Chelsea, like Genie, never really learned to speak. 00:25:14.370 --> 00:25:17.310 align:middle line:84% So once they gave her cochlear implants and she 00:25:17.310 --> 00:25:19.440 align:middle line:90% was able to hear people-- 00:25:19.440 --> 00:25:21.600 align:middle line:90% again, she learned some words. 00:25:21.600 --> 00:25:25.720 align:middle line:84% But she didn't learn how to put words together into sentences. 00:25:25.720 --> 00:25:28.580 align:middle line:84% So she learned words like "banana," and "cookie," 00:25:28.580 --> 00:25:29.080 align:middle line:90% and so on. 00:25:29.080 --> 00:25:30.360 align:middle line:84% She knew how to say those words when 00:25:30.360 --> 00:25:31.590 align:middle line:84% those were the things that she wanted 00:25:31.590 --> 00:25:33.173 align:middle line:84% or she wanted to comment on the world. 00:25:33.173 --> 00:25:36.780 align:middle line:84% But she never got past that stage. 00:25:36.780 --> 00:25:38.800 align:middle line:90% So those are two cases-- 00:25:38.800 --> 00:25:42.678 align:middle line:84% one of them maybe more convincing than the other-- 00:25:42.678 --> 00:25:44.970 align:middle line:84% that lead people to think that there might be something 00:25:44.970 --> 00:25:45.930 align:middle line:90% like a critical period. 00:25:45.930 --> 00:25:48.180 align:middle line:84% It might really be true that you're 00:25:48.180 --> 00:25:51.690 align:middle line:84% especially good at learning language in your first seven 00:25:51.690 --> 00:25:55.620 align:middle line:90% or eight years of your life. 00:25:55.620 --> 00:26:00.600 align:middle line:84% There are cases in the opposite kind. 00:26:00.600 --> 00:26:02.260 align:middle line:84% If you're in the linguistics business, 00:26:02.260 --> 00:26:04.440 align:middle line:84% you always hear about these hyperpolyglots. 00:26:04.440 --> 00:26:06.630 align:middle line:84% They're people who somehow never seem 00:26:06.630 --> 00:26:08.190 align:middle line:90% to exit their critical period. 00:26:08.190 --> 00:26:10.590 align:middle line:84% I had a professor like this, Ken Hale, 00:26:10.590 --> 00:26:13.620 align:middle line:84% who was here when I was a grad student here, 00:26:13.620 --> 00:26:16.710 align:middle line:84% somebody who seemed to just be able to learn languages 00:26:16.710 --> 00:26:20.340 align:middle line:84% by being in the same room with the speaker for a little while. 00:26:20.340 --> 00:26:23.280 align:middle line:84% He had this amazing ability to pick up languages. 00:26:23.280 --> 00:26:26.800 align:middle line:84% There are lots of Ken Hale stories, 00:26:26.800 --> 00:26:28.305 align:middle line:90% including a story about him. 00:26:28.305 --> 00:26:31.890 align:middle line:90% 00:26:31.890 --> 00:26:34.440 align:middle line:84% So he spent a big chunk of the '60s wandering 00:26:34.440 --> 00:26:37.440 align:middle line:84% around Australia, going from Aboriginal community 00:26:37.440 --> 00:26:39.690 align:middle line:84% to Aboriginal community, learning the local languages, 00:26:39.690 --> 00:26:41.860 align:middle line:84% and studying them, and writing grammars, and so on. 00:26:41.860 --> 00:26:44.490 align:middle line:84% There's at least one community that no one has 00:26:44.490 --> 00:26:46.890 align:middle line:84% been able to work with since because he 00:26:46.890 --> 00:26:48.490 align:middle line:84% was the first person who went there, 00:26:48.490 --> 00:26:50.713 align:middle line:84% and so he went and he was there for like a week, 00:26:50.713 --> 00:26:52.630 align:middle line:84% and he learned to speak the language fluently, 00:26:52.630 --> 00:26:53.338 align:middle line:90% and then he left. 00:26:53.338 --> 00:26:55.477 align:middle line:84% And then the next group of linguists 00:26:55.477 --> 00:26:57.060 align:middle line:84% came to try to work with these people. 00:26:57.060 --> 00:26:59.980 align:middle line:84% And the people were like, "You people are all so slow. 00:26:59.980 --> 00:27:01.897 align:middle line:84% Why does it take you so long to learn? 00:27:01.897 --> 00:27:04.230 align:middle line:84% We've been telling you over and over again these words." 00:27:04.230 --> 00:27:05.970 align:middle line:84% And linguists were like, "Well, it takes a while 00:27:05.970 --> 00:27:06.580 align:middle line:90% to learn the language." 00:27:06.580 --> 00:27:08.330 align:middle line:84% And they're like "No, no, no, the last guy 00:27:08.330 --> 00:27:09.490 align:middle line:90% who came through here-- 00:27:09.490 --> 00:27:11.250 align:middle line:90% it was very easy. 00:27:11.250 --> 00:27:13.440 align:middle line:84% I don't know what your problem is." 00:27:13.440 --> 00:27:14.730 align:middle line:90% Did you have a question? 00:27:14.730 --> 00:27:16.272 align:middle line:84% AUDIENCE: Yeah, is there any evidence 00:27:16.272 --> 00:27:18.765 align:middle line:84% of actual physical cerebral damage 00:27:18.765 --> 00:27:21.880 align:middle line:90% or atrophy in these cases? 00:27:21.880 --> 00:27:23.980 align:middle line:84% NORVIN RICHARDS: So for Genie, I don't 00:27:23.980 --> 00:27:26.920 align:middle line:84% know whether the relevant scans have been done. 00:27:26.920 --> 00:27:27.980 align:middle line:90% I don't know. 00:27:27.980 --> 00:27:30.443 align:middle line:84% For people like Chelsea, I also don't know. 00:27:30.443 --> 00:27:32.110 align:middle line:84% But it would be interesting to find out. 00:27:32.110 --> 00:27:34.402 align:middle line:90% Yeah, yeah. 00:27:34.402 --> 00:27:35.860 align:middle line:84% Genie-- I mean, there are all kinds 00:27:35.860 --> 00:27:39.008 align:middle line:84% of sad things about the story of Genie. 00:27:39.008 --> 00:27:40.550 align:middle line:84% After she was taken from her parents, 00:27:40.550 --> 00:27:42.133 align:middle line:84% she lived in a series of institutions. 00:27:42.133 --> 00:27:44.770 align:middle line:90% And then her mother-- 00:27:44.770 --> 00:27:46.818 align:middle line:84% her father committed suicide pretty soon 00:27:46.818 --> 00:27:47.860 align:middle line:90% after she was taken away. 00:27:47.860 --> 00:27:52.840 align:middle line:84% And her mother-- eventually, I think 00:27:52.840 --> 00:27:56.110 align:middle line:84% her mother got custody of her again for a while. 00:27:56.110 --> 00:27:58.360 align:middle line:84% And then she was back in the institutions for a while. 00:27:58.360 --> 00:28:00.610 align:middle line:84% So she lived for a while with some linguists 00:28:00.610 --> 00:28:05.980 align:middle line:84% who wrote books about her, studied her, tried to help her. 00:28:05.980 --> 00:28:10.510 align:middle line:84% And then her mother successfully convinced the court 00:28:10.510 --> 00:28:13.650 align:middle line:84% that she would be better off living with her mother. 00:28:13.650 --> 00:28:15.555 align:middle line:90% So they did that for a while. 00:28:15.555 --> 00:28:16.930 align:middle line:84% So there was a period there where 00:28:16.930 --> 00:28:21.040 align:middle line:84% she was hidden from the eyes of linguists 00:28:21.040 --> 00:28:22.753 align:middle line:84% after this initial period, where she 00:28:22.753 --> 00:28:23.920 align:middle line:90% was being intensely studied. 00:28:23.920 --> 00:28:28.260 align:middle line:90% 00:28:28.260 --> 00:28:32.160 align:middle line:84% OK, all right, so it's possible anyway 00:28:32.160 --> 00:28:34.632 align:middle line:84% that there's a stage that children 00:28:34.632 --> 00:28:36.840 align:middle line:84% go through where they're especially good at acquiring 00:28:36.840 --> 00:28:39.540 align:middle line:84% language, a stage that most people then exit around six 00:28:39.540 --> 00:28:41.370 align:middle line:84% or seven years of age, though there 00:28:41.370 --> 00:28:45.150 align:middle line:84% might be people like my old Professor Ken Hale for whom 00:28:45.150 --> 00:28:46.950 align:middle line:90% it just never ends. 00:28:46.950 --> 00:28:50.370 align:middle line:84% And then lots of questions about how this knowledge works-- 00:28:50.370 --> 00:28:52.140 align:middle line:90% how do you acquire it? 00:28:52.140 --> 00:28:53.640 align:middle line:84% And we've been spending the semester 00:28:53.640 --> 00:28:55.590 align:middle line:84% talking about the nature of this knowledge. 00:28:55.590 --> 00:28:57.090 align:middle line:84% There are people who specialize in what's 00:28:57.090 --> 00:28:58.257 align:middle line:90% called language acquisition. 00:28:58.257 --> 00:29:02.260 align:middle line:84% This is the study of what kids know and when they know it. 00:29:02.260 --> 00:29:04.350 align:middle line:84% And so I wanted to tell you a few of the results 00:29:04.350 --> 00:29:05.820 align:middle line:90% from that subdomain. 00:29:05.820 --> 00:29:09.180 align:middle line:84% There is an undergraduate class that specializes 00:29:09.180 --> 00:29:10.320 align:middle line:90% in language acquisition. 00:29:10.320 --> 00:29:12.580 align:middle line:90% I forget what its number is. 00:29:12.580 --> 00:29:16.200 align:middle line:84% But if this is a series of topics that interests you, 00:29:16.200 --> 00:29:19.560 align:middle line:84% we have people who can teach you about this. 00:29:19.560 --> 00:29:22.050 align:middle line:84% There is a frequent discovery that the people who 00:29:22.050 --> 00:29:23.790 align:middle line:84% do language acquisition make, which 00:29:23.790 --> 00:29:26.440 align:middle line:84% is that children are smarter than you might think. 00:29:26.440 --> 00:29:30.180 align:middle line:84% So they know stuff that it's not obvious 00:29:30.180 --> 00:29:34.560 align:middle line:84% that they know just by watching them. 00:29:34.560 --> 00:29:38.678 align:middle line:84% Lots of cases where children are making what we may as well call 00:29:38.678 --> 00:29:40.470 align:middle line:84% an error-- that is, they're doing something 00:29:40.470 --> 00:29:42.040 align:middle line:90% that adults don't do. 00:29:42.040 --> 00:29:45.360 align:middle line:84% But if you look harder, you can show that they actually 00:29:45.360 --> 00:29:48.090 align:middle line:84% do have a representation, some knowledge that 00:29:48.090 --> 00:29:50.010 align:middle line:90% resembles what adults do. 00:29:50.010 --> 00:29:52.830 align:middle line:84% I'll show you a phonological example first. 00:29:52.830 --> 00:29:57.870 align:middle line:84% Here are some data from a Japanese child. 00:29:57.870 --> 00:29:59.460 align:middle line:84% This is a standard representation 00:29:59.460 --> 00:30:00.420 align:middle line:90% of children's ages. 00:30:00.420 --> 00:30:02.453 align:middle line:84% So this Japanese child, at the time 00:30:02.453 --> 00:30:03.870 align:middle line:84% that these observations were made, 00:30:03.870 --> 00:30:06.130 align:middle line:84% was three years and two months old. 00:30:06.130 --> 00:30:09.700 align:middle line:84% That's what that 3:2 thing means. 00:30:09.700 --> 00:30:11.640 align:middle line:84% So observation about this Japanese child-- 00:30:11.640 --> 00:30:14.970 align:middle line:84% this Japanese child didn't have the velar 00:30:14.970 --> 00:30:16.830 align:middle line:90% stop that Japanese has. 00:30:16.830 --> 00:30:22.740 align:middle line:84% So they were changing the velar stops to alveolar stops-- 00:30:22.740 --> 00:30:23.550 align:middle line:90% to "t"s, right? 00:30:23.550 --> 00:30:25.630 align:middle line:90% So mikaN, which is an orange-- 00:30:25.630 --> 00:30:28.650 align:middle line:84% it's not exactly an orange, it's another kind of citrus fruit-- 00:30:28.650 --> 00:30:30.480 align:middle line:90% is "mitaN" for this child. 00:30:30.480 --> 00:30:32.610 align:middle line:84% Or "poketto," which is a borrowing from English 00:30:32.610 --> 00:30:34.320 align:middle line:90% "pocket," is "potetto." 00:30:34.320 --> 00:30:36.770 align:middle line:84% Or "neko," which is "cat," becomes "neto." 00:30:36.770 --> 00:30:41.950 align:middle line:84% Yeah, so this child is changing "k" to "t." 00:30:41.950 --> 00:30:43.180 align:middle line:90% They also have "t." 00:30:43.180 --> 00:30:46.690 align:middle line:84% So Japanese words that have "t" in them, 00:30:46.690 --> 00:30:48.460 align:middle line:84% like "tama," which is "ball," or "terebi," 00:30:48.460 --> 00:30:51.910 align:middle line:84% which is "television"-- it's another borrowing-- 00:30:51.910 --> 00:30:52.990 align:middle line:90% it's a clipped borrowing. 00:30:52.990 --> 00:30:55.060 align:middle line:90% They do that a lot. 00:30:55.060 --> 00:30:56.680 align:middle line:90% They have "t" in those, too. 00:30:56.680 --> 00:30:58.720 align:middle line:84% So this is a child who has changed "k" to "t" 00:30:58.720 --> 00:31:01.060 align:middle line:90% and also has "t." 00:31:01.060 --> 00:31:03.610 align:middle line:84% Now, here's the thing about "t" in Japanese. 00:31:03.610 --> 00:31:06.790 align:middle line:84% "t" in Japanese undergoes some phonological changes, 00:31:06.790 --> 00:31:08.260 align:middle line:84% depending on the following vowel. 00:31:08.260 --> 00:31:13.270 align:middle line:84% So if "t" is followed, for example, by the vowel "a," 00:31:13.270 --> 00:31:14.320 align:middle line:90% it's just a "t." 00:31:14.320 --> 00:31:17.800 align:middle line:90% So "doesn't wait" is "matanai." 00:31:17.800 --> 00:31:19.720 align:middle line:84% But if you put that verb in the positive form, 00:31:19.720 --> 00:31:21.460 align:middle line:84% then the vowel that follows the "t" 00:31:21.460 --> 00:31:23.440 align:middle line:84% won't be an "a," it'll be an "u"-- 00:31:23.440 --> 00:31:25.810 align:middle line:90% the high back unrounded vowel. 00:31:25.810 --> 00:31:28.870 align:middle line:84% And before "u," "t" changes to "tsu." 00:31:28.870 --> 00:31:30.830 align:middle line:90% That's just a general rule. 00:31:30.830 --> 00:31:32.470 align:middle line:84% If you ever learn Japanese, you'll 00:31:32.470 --> 00:31:36.250 align:middle line:84% see it's encoded in their writing system, 00:31:36.250 --> 00:31:40.720 align:middle line:84% in the kana, that a "t" before an "u" is a "tsu." 00:31:40.720 --> 00:31:44.335 align:middle line:84% So "waits," the positive version of "matanai," is "matsu." 00:31:44.335 --> 00:31:47.150 align:middle line:90% The "t" changes to a "tsu." 00:31:47.150 --> 00:31:50.080 align:middle line:84% And if "t" is before the vowel "i," 00:31:50.080 --> 00:31:52.840 align:middle line:84% the high front vowel, it [INAUDIBLE] to a "ch." 00:31:52.840 --> 00:31:55.690 align:middle line:84% So the word for "city," for example, is "matchi," yeah, 00:31:55.690 --> 00:31:59.830 align:middle line:84% or the formal version of "waits" is "matchimas," 00:31:59.830 --> 00:32:02.330 align:middle line:84% so you add "imas" to "mat-," that stem "mat-," 00:32:02.330 --> 00:32:03.460 align:middle line:90% which means "wait." 00:32:03.460 --> 00:32:04.930 align:middle line:84% You get the "t" changing to a "ch." 00:32:04.930 --> 00:32:09.143 align:middle line:90% OK, so basic Japanese phonology. 00:32:09.143 --> 00:32:10.060 align:middle line:90% Now, here's the thing. 00:32:10.060 --> 00:32:12.640 align:middle line:84% This child had all of that basic Japanese phonology. 00:32:12.640 --> 00:32:14.860 align:middle line:84% So that child was saying, at three years 00:32:14.860 --> 00:32:18.400 align:middle line:84% and two months, "matanai," "matsu," and "matchi." 00:32:18.400 --> 00:32:22.180 align:middle line:84% But this child was also saying, for autumn, "ati." 00:32:22.180 --> 00:32:25.840 align:middle line:84% That is, the child was changing "k" to "t." 00:32:25.840 --> 00:32:28.720 align:middle line:84% And the child changing "k" to "t"- 00:32:28.720 --> 00:32:31.660 align:middle line:84% so the adult word for autumn is "aki." 00:32:31.660 --> 00:32:34.270 align:middle line:84% And the adult word for "bear" is "kuma." 00:32:34.270 --> 00:32:38.950 align:middle line:84% And this child knows that "t" becomes "ch" before "i," right? 00:32:38.950 --> 00:32:42.310 align:middle line:84% But autumn, for this child, is not "achi." 00:32:42.310 --> 00:32:43.780 align:middle line:90% It's "ati," right? 00:32:43.780 --> 00:32:48.040 align:middle line:84% And "bear" for this child is not "tsuma," where "t" normally 00:32:48.040 --> 00:32:51.760 align:middle line:84% becomes "tsu" before "u" like it does in "wait." 00:32:51.760 --> 00:32:54.010 align:middle line:84% For this child, the word for "bear" is not "tsuma." 00:32:54.010 --> 00:32:56.320 align:middle line:90% It's "tuma," yeah? 00:32:56.320 --> 00:32:58.462 align:middle line:84% If we were doing this in terms of rule ordering, 00:32:58.462 --> 00:32:59.920 align:middle line:84% I guess we could say, yeah, this is 00:32:59.920 --> 00:33:04.840 align:middle line:84% a child for whom there is a distinction between "t" 00:33:04.840 --> 00:33:05.740 align:middle line:90% and "k." 00:33:05.740 --> 00:33:07.570 align:middle line:90% And then there's a set of rules. 00:33:07.570 --> 00:33:10.240 align:middle line:84% There are the rules that change "t" to "tsu" or "ch" 00:33:10.240 --> 00:33:11.740 align:middle line:90% before certain vowels, right? 00:33:11.740 --> 00:33:13.900 align:middle line:84% And then after that, there's another rule that 00:33:13.900 --> 00:33:16.330 align:middle line:90% changes "k" to "t." 00:33:16.330 --> 00:33:19.528 align:middle line:84% Your first thought, looking at the first set I did, "orange," 00:33:19.528 --> 00:33:21.070 align:middle line:84% and "pocket," and "cat" would be, oh, 00:33:21.070 --> 00:33:25.730 align:middle line:84% how cute this child can't say "k," and they say "t" instead. 00:33:25.730 --> 00:33:28.960 align:middle line:84% And that's right in a sense, right? 00:33:28.960 --> 00:33:30.550 align:middle line:90% You might have thought then-- 00:33:30.550 --> 00:33:33.040 align:middle line:84% this is a child who-- their Japanese 00:33:33.040 --> 00:33:36.520 align:middle line:84% doesn't have as many sounds in it as adult Japanese. 00:33:36.520 --> 00:33:38.847 align:middle line:84% And that's also right, in a sense. 00:33:38.847 --> 00:33:40.930 align:middle line:84% But there's another sense in which it's not right. 00:33:40.930 --> 00:33:44.260 align:middle line:84% The child does distinguish "t" and "k," right? 00:33:44.260 --> 00:33:47.800 align:middle line:84% It's just that they pronounce them both "t." 00:33:47.800 --> 00:33:51.010 align:middle line:84% So the "t" that's really a "t" is 00:33:51.010 --> 00:33:53.560 align:middle line:84% the one that undergoes these sound changes before vowels. 00:33:53.560 --> 00:33:55.873 align:middle line:84% And the "t" that comes from "k" is just always a "t." 00:33:55.873 --> 00:33:57.790 align:middle line:84% It doesn't matter what the following vowel is. 00:33:57.790 --> 00:34:00.760 align:middle line:84% Yeah, so there's some sense in which this child has 00:34:00.760 --> 00:34:02.200 align:middle line:90% a full adult representation. 00:34:02.200 --> 00:34:05.890 align:middle line:84% You have to imagine them having "aki" in their head, right? 00:34:05.890 --> 00:34:08.800 align:middle line:84% And it's just that there is some rule that 00:34:08.800 --> 00:34:11.199 align:middle line:84% goes between their head and their lips 00:34:11.199 --> 00:34:13.517 align:middle line:84% that changes "k" to "t" all the time, 00:34:13.517 --> 00:34:15.100 align:middle line:84% no matter what the following vowel is. 00:34:15.100 --> 00:34:17.170 align:middle line:84% There's something like that going on. 00:34:17.170 --> 00:34:20.739 align:middle line:84% There's a small instance of a general pattern. 00:34:20.739 --> 00:34:25.389 align:middle line:84% What you find is kids are doing something wrong. 00:34:25.389 --> 00:34:28.340 align:middle line:84% They're doing something that's not what adults do. 00:34:28.340 --> 00:34:30.130 align:middle line:84% But if you look very closely, you 00:34:30.130 --> 00:34:32.830 align:middle line:84% can convince yourself that they know more than they seem to. 00:34:32.830 --> 00:34:38.139 align:middle line:90% 00:34:38.139 --> 00:34:41.800 align:middle line:84% Yeah, so just to remind you of stuff that we 00:34:41.800 --> 00:34:43.850 align:middle line:90% talked about earlier on-- 00:34:43.850 --> 00:34:46.900 align:middle line:84% so what I just showed you-- and I'll show you 00:34:46.900 --> 00:34:49.389 align:middle line:84% some more examples like this in a second-- 00:34:49.389 --> 00:34:52.600 align:middle line:84% involve just making lots of recordings 00:34:52.600 --> 00:34:55.480 align:middle line:84% of children speaking naturally, and then studying them. 00:34:55.480 --> 00:34:59.890 align:middle line:84% And that is one kind of thing that acquisitionists do. 00:34:59.890 --> 00:35:03.840 align:middle line:84% There are other kinds of work in which you 00:35:03.840 --> 00:35:06.350 align:middle line:90% run experiments on children. 00:35:06.350 --> 00:35:08.410 align:middle line:84% We talked about some of this kind of work 00:35:08.410 --> 00:35:09.860 align:middle line:90% early on in the class. 00:35:09.860 --> 00:35:13.030 align:middle line:84% So we're talking about selective looking tasks 00:35:13.030 --> 00:35:15.250 align:middle line:90% and sucking rate tasks. 00:35:15.250 --> 00:35:19.240 align:middle line:84% So sucking rate tasks make use of the fact 00:35:19.240 --> 00:35:22.530 align:middle line:84% that if you give a very small infant something to suck on-- 00:35:22.530 --> 00:35:23.530 align:middle line:90% first of all, they will. 00:35:23.530 --> 00:35:25.572 align:middle line:84% Sucking on things is one of their favorite things 00:35:25.572 --> 00:35:26.650 align:middle line:90% to do at that age. 00:35:26.650 --> 00:35:29.980 align:middle line:84% And it turns out that if you give them something to suck on, 00:35:29.980 --> 00:35:32.080 align:middle line:84% they will suck on it at first fairly rapidly. 00:35:32.080 --> 00:35:34.030 align:middle line:84% And then they will start to slow down. 00:35:34.030 --> 00:35:38.290 align:middle line:84% And how quickly-- they say you can give them something to suck 00:35:38.290 --> 00:35:42.550 align:middle line:84% on to which you have attached a clever device that will record 00:35:42.550 --> 00:35:45.460 align:middle line:84% the rate at which they suck, pay attention to how often 00:35:45.460 --> 00:35:46.150 align:middle line:90% they're going-- 00:35:46.150 --> 00:35:49.090 align:middle line:90% they're making sucking motions. 00:35:49.090 --> 00:35:51.340 align:middle line:84% And the observation is that if you 00:35:51.340 --> 00:35:53.485 align:middle line:84% give a child something interesting to look at, 00:35:53.485 --> 00:35:55.360 align:middle line:84% if they've started to slow down, and then you 00:35:55.360 --> 00:35:56.440 align:middle line:84% show them an interesting picture, 00:35:56.440 --> 00:35:57.790 align:middle line:90% they'll speed back up again. 00:35:57.790 --> 00:35:59.440 align:middle line:90% They perk up. 00:35:59.440 --> 00:36:02.720 align:middle line:84% And language acquisition is to make use of this. 00:36:02.720 --> 00:36:04.390 align:middle line:90% So if you want to know-- 00:36:04.390 --> 00:36:06.490 align:middle line:84% so here's the kind of experiment people have done. 00:36:06.490 --> 00:36:09.170 align:middle line:84% They've shown that Japanese infants-- 00:36:09.170 --> 00:36:11.110 align:middle line:84% so very small Japanese children-- 00:36:11.110 --> 00:36:14.740 align:middle line:84% can distinguish an "r" sound from an "l" sound. 00:36:14.740 --> 00:36:18.540 align:middle line:84% That is, if you play "rah" or "lah" at them, 00:36:18.540 --> 00:36:20.540 align:middle line:84% that they can distinguish these from each other. 00:36:20.540 --> 00:36:22.665 align:middle line:84% This is interesting because adult Japanese speakers 00:36:22.665 --> 00:36:24.190 align:middle line:90% can't do that. 00:36:24.190 --> 00:36:25.840 align:middle line:84% Japanese doesn't have a distinction 00:36:25.840 --> 00:36:27.280 align:middle line:90% between "r" and "l." 00:36:27.280 --> 00:36:29.020 align:middle line:90% They only have one liquid. 00:36:29.020 --> 00:36:32.830 align:middle line:84% So it's a famous fact about Japanese speakers. 00:36:32.830 --> 00:36:35.470 align:middle line:84% It's hard for them to distinguish "r" from "l." 00:36:35.470 --> 00:36:37.390 align:middle line:84% But very small Japanese children-- 00:36:37.390 --> 00:36:39.580 align:middle line:84% so children who are acquiring Japanese-- 00:36:39.580 --> 00:36:40.838 align:middle line:90% can distinguish "r" from "l." 00:36:40.838 --> 00:36:42.880 align:middle line:84% They know the difference between "rah" and "lah." 00:36:42.880 --> 00:36:46.420 align:middle line:84% And the way you show that is with this sucking rate task. 00:36:46.420 --> 00:36:48.285 align:middle line:84% So you give the child something to suck on. 00:36:48.285 --> 00:36:49.660 align:middle line:84% And when they start to slow down, 00:36:49.660 --> 00:36:53.950 align:middle line:84% you start playing "rah, rah, rah, rah" at them. 00:36:53.950 --> 00:36:56.410 align:middle line:84% And they speed up at first because, hey, here's 00:36:56.410 --> 00:36:57.400 align:middle line:90% a new thing. 00:36:57.400 --> 00:36:58.180 align:middle line:90% It's a recording. 00:36:58.180 --> 00:36:59.600 align:middle line:84% And then eventually, they slow down. 00:36:59.600 --> 00:37:01.017 align:middle line:84% They're like "rah," yes, I get it. 00:37:01.017 --> 00:37:02.170 align:middle line:90% Thank you very much. 00:37:02.170 --> 00:37:04.478 align:middle line:90% And then you switch to "lah." 00:37:04.478 --> 00:37:06.770 align:middle line:84% So you go "rah, rah, rah," and they start to slow down. 00:37:06.770 --> 00:37:11.020 align:middle line:84% Then you go "lah, lah, lah," and they perk back up again, 00:37:11.020 --> 00:37:12.490 align:middle line:84% showing that they can distinguish 00:37:12.490 --> 00:37:14.865 align:middle line:84% the "rah" from the "lah," that these are different things 00:37:14.865 --> 00:37:17.200 align:middle line:90% as far as they're concerned. 00:37:17.200 --> 00:37:18.700 align:middle line:84% They've done similar kinds of tests 00:37:18.700 --> 00:37:23.530 align:middle line:84% with English-acquiring children, showing 00:37:23.530 --> 00:37:25.630 align:middle line:84% that they can distinguish aspirated 00:37:25.630 --> 00:37:27.310 align:middle line:90% from unaspirated stops. 00:37:27.310 --> 00:37:30.080 align:middle line:84% So cast your minds back to the beginning of the semester. 00:37:30.080 --> 00:37:33.010 align:middle line:84% We were talking about the fact that in a language like Hindi, 00:37:33.010 --> 00:37:34.450 align:middle line:84% for example, there's a difference 00:37:34.450 --> 00:37:36.820 align:middle line:84% between "pah" [ASPIRATED] and "pah" [UNASPIRATED].. 00:37:36.820 --> 00:37:41.120 align:middle line:84% So English has voiced stops and voiceless stops. 00:37:41.120 --> 00:37:46.000 align:middle line:84% We have a distinction between "pah" and "bah." 00:37:46.000 --> 00:37:48.130 align:middle line:84% So there's an aspirated voiceless stop. 00:37:48.130 --> 00:37:49.960 align:middle line:90% And there's a voiced stop. 00:37:49.960 --> 00:37:52.690 align:middle line:84% And then Hindi has a three-way distinction 00:37:52.690 --> 00:37:57.400 align:middle line:84% between "pah" [ASPIRATED] and "pah" [UNASPIRATED] and "bah." 00:37:57.400 --> 00:38:00.892 align:middle line:84% And so "pahl" and "pahl" are two different words in Hindi. 00:38:00.892 --> 00:38:03.100 align:middle line:84% I think one of them means "knife" and the other means 00:38:03.100 --> 00:38:03.640 align:middle line:90% "flour." 00:38:03.640 --> 00:38:05.140 align:middle line:84% And I can't remember which is which. 00:38:05.140 --> 00:38:07.700 align:middle line:90% Anybody here speak Hindi? 00:38:07.700 --> 00:38:09.240 align:middle line:84% OK, those are two different words. 00:38:09.240 --> 00:38:16.230 align:middle line:84% So Hindi has these two distinct sounds that are different. 00:38:16.230 --> 00:38:19.280 align:middle line:84% And an observation that people have 00:38:19.280 --> 00:38:22.370 align:middle line:84% made with very small infants who are acquiring English 00:38:22.370 --> 00:38:24.740 align:middle line:84% is that if you do the sucking rate task with them 00:38:24.740 --> 00:38:28.670 align:middle line:84% and you play "pah, pah, pah" at them, that they eventually 00:38:28.670 --> 00:38:29.730 align:middle line:90% start to slow down. 00:38:29.730 --> 00:38:32.171 align:middle line:84% And if you then switch to "pah, pah, pah," 00:38:32.171 --> 00:38:33.920 align:middle line:84% the unaspirated stop, they're like, 00:38:33.920 --> 00:38:36.350 align:middle line:84% oh yeah, that's a different thing. 00:38:36.350 --> 00:38:38.990 align:middle line:84% So they recognize that as a different sound. 00:38:38.990 --> 00:38:41.330 align:middle line:84% One of the things that they can show-- 00:38:41.330 --> 00:38:43.880 align:middle line:84% actually, because the beauty of aspiration 00:38:43.880 --> 00:38:48.770 align:middle line:84% is it's the kind of thing that comes in degrees that with-- 00:38:48.770 --> 00:38:53.000 align:middle line:84% that you have the release of the closure, and then 00:38:53.000 --> 00:38:54.620 align:middle line:84% you have the beginning of the vowel. 00:38:54.620 --> 00:38:57.140 align:middle line:84% And you can electronically manipulate 00:38:57.140 --> 00:39:00.710 align:middle line:84% exactly how many milliseconds it takes for the vowel to start. 00:39:00.710 --> 00:39:02.760 align:middle line:84% And there is a particular cutoff. 00:39:02.760 --> 00:39:07.250 align:middle line:84% So if you give children an aspirated stop 00:39:07.250 --> 00:39:10.010 align:middle line:84% and an unaspirated stop, they can distinguish those. 00:39:10.010 --> 00:39:13.910 align:middle line:84% If you give them two stops that differ in aspiration by just 00:39:13.910 --> 00:39:17.690 align:middle line:84% a little bit, then whether they can distinguish those two 00:39:17.690 --> 00:39:20.570 align:middle line:84% stops from each other depends on whether one of them 00:39:20.570 --> 00:39:23.210 align:middle line:84% is on one side and the other on the other side 00:39:23.210 --> 00:39:24.890 align:middle line:90% of a particular dividing line. 00:39:24.890 --> 00:39:27.140 align:middle line:84% I forget what it is, like 50 milliseconds or something 00:39:27.140 --> 00:39:27.640 align:middle line:90% like that. 00:39:27.640 --> 00:39:29.360 align:middle line:84% There's some particular amount of time 00:39:29.360 --> 00:39:31.790 align:middle line:84% that is the cutoff for distinguishing sounds 00:39:31.790 --> 00:39:33.440 align:middle line:90% from each other. 00:39:33.440 --> 00:39:37.610 align:middle line:84% For being able to speak adult English, 00:39:37.610 --> 00:39:39.620 align:middle line:90% this knowledge is useless. 00:39:39.620 --> 00:39:43.190 align:middle line:84% But the theory is that all babies-- 00:39:43.190 --> 00:39:46.340 align:middle line:84% Japanese babies are all ready to distinguish "l" and "r." 00:39:46.340 --> 00:39:48.305 align:middle line:84% And the English speaking babies are all 00:39:48.305 --> 00:39:51.260 align:middle line:84% ready to distinguish aspirated and unaspirated stops. 00:39:51.260 --> 00:39:53.480 align:middle line:84% The babies are just born able to distinguish 00:39:53.480 --> 00:39:55.710 align:middle line:84% all of the kinds of sounds that they might need. 00:39:55.710 --> 00:39:57.260 align:middle line:84% And then as they grow up and start 00:39:57.260 --> 00:39:59.385 align:middle line:84% learning the actual language that they're learning, 00:39:59.385 --> 00:40:01.070 align:middle line:90% they forget stuff. 00:40:01.070 --> 00:40:03.260 align:middle line:84% Anything that isn't useful they get rid of. 00:40:03.260 --> 00:40:06.410 align:middle line:84% So Japanese babies stop carefully distinguishing "rah" 00:40:06.410 --> 00:40:10.154 align:middle line:84% from "lah" because their parents don't. 00:40:10.154 --> 00:40:12.285 align:middle line:84% That's one of the observations people 00:40:12.285 --> 00:40:13.410 align:middle line:90% have made about this stuff. 00:40:13.410 --> 00:40:16.750 align:middle line:90% 00:40:16.750 --> 00:40:20.832 align:middle line:84% So I want to get on to the next thing. 00:40:20.832 --> 00:40:22.790 align:middle line:84% So I don't want to spend too much time on this. 00:40:22.790 --> 00:40:25.580 align:middle line:90% 00:40:25.580 --> 00:40:27.560 align:middle line:84% So we've spent some time talking about stuff 00:40:27.560 --> 00:40:29.870 align:middle line:84% that suggests that babies have just 00:40:29.870 --> 00:40:33.020 align:middle line:84% a spooky amount of knowledge in advance, 00:40:33.020 --> 00:40:34.970 align:middle line:84% before they learn anything at all. 00:40:34.970 --> 00:40:39.170 align:middle line:84% And in fact, in the particular case of acquiring phonology, 00:40:39.170 --> 00:40:41.420 align:middle line:84% some of what they're doing is forgetting stuff 00:40:41.420 --> 00:40:43.010 align:middle line:84% that they come into the world knowing. 00:40:43.010 --> 00:40:45.170 align:middle line:84% That seems to be the way they're set up. 00:40:45.170 --> 00:40:51.320 align:middle line:84% There is stuff that children clearly have to compute. 00:40:51.320 --> 00:40:54.470 align:middle line:84% And there's some beautiful work on how exactly 00:40:54.470 --> 00:40:56.960 align:middle line:84% children segment the speech stream. 00:40:56.960 --> 00:41:00.260 align:middle line:84% So when we write things in English, 00:41:00.260 --> 00:41:03.170 align:middle line:84% we write with spaces in between words. 00:41:03.170 --> 00:41:04.970 align:middle line:84% But those spaces don't correspond 00:41:04.970 --> 00:41:07.070 align:middle line:84% to anything in the phonological signal, right? 00:41:07.070 --> 00:41:09.950 align:middle line:84% If you look at a spectrogram of somebody speaking, 00:41:09.950 --> 00:41:12.380 align:middle line:84% it's just a long string of sound. 00:41:12.380 --> 00:41:14.510 align:middle line:84% You don't put spaces in between your words, 00:41:14.510 --> 00:41:18.890 align:middle line:84% unless you're speaking extremely cautiously. 00:41:18.890 --> 00:41:21.050 align:middle line:84% And in fact, children make mistakes 00:41:21.050 --> 00:41:22.070 align:middle line:90% in speech segmentation. 00:41:22.070 --> 00:41:26.000 align:middle line:84% So a classic example is of a child who's 00:41:26.000 --> 00:41:29.880 align:middle line:84% just been told to behave and responds, "I am being hayve." 00:41:29.880 --> 00:41:33.770 align:middle line:84% So this is a child who thinks that "behave" is like "be" 00:41:33.770 --> 00:41:36.380 align:middle line:84% plus some adjective, right, "hayve"? 00:41:36.380 --> 00:41:38.138 align:middle line:90% "Be as hayve as you can," right? 00:41:38.138 --> 00:41:39.680 align:middle line:84% The child is doing their best, right? 00:41:39.680 --> 00:41:42.290 align:middle line:84% Or the other classic like this is 00:41:42.290 --> 00:41:46.340 align:middle line:84% a child who's just been told "We're going to go to Miami" 00:41:46.340 --> 00:41:50.090 align:middle line:84% and responds "I don't want to go to your Ami," 00:41:50.090 --> 00:41:51.950 align:middle line:84% which is a similar kind of mistake, 00:41:51.950 --> 00:41:54.620 align:middle line:84% is thinking that "Miami" is two words, a perfectly plausible 00:41:54.620 --> 00:41:56.760 align:middle line:90% thing. 00:41:56.760 --> 00:41:59.960 align:middle line:84% There's all kinds of really interesting work on how exactly 00:41:59.960 --> 00:42:02.150 align:middle line:84% children segment the speech stream. 00:42:02.150 --> 00:42:07.100 align:middle line:84% And there's lots of evidence that they use statistics about 00:42:07.100 --> 00:42:12.710 align:middle line:84% phonotactics-- that is, they make observations like if I see 00:42:12.710 --> 00:42:17.870 align:middle line:84% this pair of sounds next to each other, it's because-- 00:42:17.870 --> 00:42:23.960 align:middle line:84% so in English, if I have an expression-- 00:42:23.960 --> 00:42:30.440 align:middle line:84% if I say a pair of words like "lies beneath" the water, 00:42:30.440 --> 00:42:31.790 align:middle line:90% let's say-- 00:42:31.790 --> 00:42:34.850 align:middle line:84% so here, I've got a word ending in a "z" and another word 00:42:34.850 --> 00:42:37.040 align:middle line:90% beginning with a "b," yeah? 00:42:37.040 --> 00:42:39.290 align:middle line:84% So if you're an English speaking child, 00:42:39.290 --> 00:42:42.050 align:middle line:84% even before you know any words, the observation 00:42:42.050 --> 00:42:44.810 align:middle line:84% is-- it's as though they look at pairs like that 00:42:44.810 --> 00:42:46.970 align:middle line:84% and they go, oh, I don't ever seem 00:42:46.970 --> 00:42:51.500 align:middle line:84% to hear speech streams beginning with "zb" in English. 00:42:51.500 --> 00:42:55.560 align:middle line:84% So "zb" isn't something that an English word can begin with. 00:42:55.560 --> 00:42:58.010 align:middle line:84% And so maybe this is a word boundary. 00:42:58.010 --> 00:43:04.220 align:middle line:84% And there's this experiment that Saffran et al. did in '96 00:43:04.220 --> 00:43:06.680 align:middle line:84% showing that very young children are 00:43:06.680 --> 00:43:10.610 align:middle line:84% very good at-- this was, again, using the sucking task, 00:43:10.610 --> 00:43:11.150 align:middle line:90% I think. 00:43:11.150 --> 00:43:13.460 align:middle line:84% Maybe it was the selective looking task, a similar kind 00:43:13.460 --> 00:43:16.910 align:middle line:84% of task, showing the children are 00:43:16.910 --> 00:43:23.030 align:middle line:84% very good at making observations about which pairs of sounds 00:43:23.030 --> 00:43:25.460 align:middle line:84% occur next to each other frequently in a speech stream. 00:43:25.460 --> 00:43:27.710 align:middle line:84% So in their particular case, what they were doing 00:43:27.710 --> 00:43:31.880 align:middle line:84% was exposing small children to streams of syllables. 00:43:31.880 --> 00:43:34.070 align:middle line:90% So they made up words. 00:43:34.070 --> 00:43:35.452 align:middle line:90% I'm going to make up words, too. 00:43:35.452 --> 00:43:37.160 align:middle line:84% So this is not the actual words they had, 00:43:37.160 --> 00:43:38.118 align:middle line:90% but they had the shape. 00:43:38.118 --> 00:43:40.710 align:middle line:90% They were CV, CV, CV words. 00:43:40.710 --> 00:43:49.970 align:middle line:84% So it was like "to, la, vi," and then "mu, ga, pi," whatever. 00:43:49.970 --> 00:43:53.270 align:middle line:84% And they made up strings of words. 00:43:53.270 --> 00:43:56.350 align:middle line:90% Here, I'll make up one more-- 00:43:56.350 --> 00:44:00.080 align:middle line:90% "ri, na, vi." 00:44:00.080 --> 00:44:02.270 align:middle line:90% And the observation was-- 00:44:02.270 --> 00:44:07.280 align:middle line:84% so they made up speech streams for the children 00:44:07.280 --> 00:44:08.570 align:middle line:90% that involved these words. 00:44:08.570 --> 00:44:12.440 align:middle line:84% So the speech streams would sound like "to la bi, mu ga bi, 00:44:12.440 --> 00:44:15.130 align:middle line:84% re na vi, re na vi, mu ga pi, re na vi, re a vi, mu ga pi, 00:44:15.130 --> 00:44:18.650 align:middle line:90% to la bi," without intonation. 00:44:18.650 --> 00:44:21.140 align:middle line:84% Just these syllables one after another. 00:44:21.140 --> 00:44:23.840 align:middle line:84% And the children basically very quickly got 00:44:23.840 --> 00:44:26.390 align:middle line:84% to where they knew that if you hear a "to," 00:44:26.390 --> 00:44:28.640 align:middle line:84% you're going to hear a "la" next. 00:44:28.640 --> 00:44:31.070 align:middle line:84% Or if you hear a "ga," you're going to hear a "pi" next. 00:44:31.070 --> 00:44:34.940 align:middle line:84% So they got to where they could make those predictions. 00:44:34.940 --> 00:44:38.540 align:middle line:84% And what that meant was that they knew that if you 00:44:38.540 --> 00:44:40.130 align:middle line:84% heard a "pi," you didn't know what 00:44:40.130 --> 00:44:41.297 align:middle line:90% you were going to hear next. 00:44:41.297 --> 00:44:43.910 align:middle line:84% There were a variety of things that you could hear next. 00:44:43.910 --> 00:44:46.760 align:middle line:84% So they learned these transitional probabilities 00:44:46.760 --> 00:44:47.405 align:middle line:90% very quickly. 00:44:47.405 --> 00:44:49.670 align:middle line:90% And it was remarkable. 00:44:49.670 --> 00:44:51.500 align:middle line:84% Again, babies are smarter than they look. 00:44:51.500 --> 00:44:55.820 align:middle line:84% This is one of the big results of language acquisition 00:44:55.820 --> 00:44:56.480 align:middle line:90% research. 00:44:56.480 --> 00:44:59.660 align:middle line:90% So children are good at this. 00:44:59.660 --> 00:45:02.850 align:middle line:84% There's a lot of work since '96 on exactly what it is they're 00:45:02.850 --> 00:45:03.350 align:middle line:90% good at. 00:45:03.350 --> 00:45:05.330 align:middle line:84% So for one thing, the original work 00:45:05.330 --> 00:45:07.310 align:middle line:84% involved words, made up words that 00:45:07.310 --> 00:45:08.840 align:middle line:90% were all three syllables long. 00:45:08.840 --> 00:45:11.340 align:middle line:84% And it was shown that if you'd vary the number of syllables, 00:45:11.340 --> 00:45:12.838 align:middle line:90% the babies get less good at it. 00:45:12.838 --> 00:45:15.380 align:middle line:84% It was also shown that they get better at it if you give them 00:45:15.380 --> 00:45:17.150 align:middle line:84% prosodic cues, like the words are all 00:45:17.150 --> 00:45:20.000 align:middle line:84% stressed on the first syllable, that that helps figure out 00:45:20.000 --> 00:45:23.660 align:middle line:90% where the word boundaries are. 00:45:23.660 --> 00:45:26.120 align:middle line:84% So however they're doing it, the upshot 00:45:26.120 --> 00:45:27.770 align:middle line:90% of it all is that children-- 00:45:27.770 --> 00:45:29.843 align:middle line:84% they speak their first words around the end 00:45:29.843 --> 00:45:31.010 align:middle line:90% of their first year of life. 00:45:31.010 --> 00:45:33.405 align:middle line:84% And then by the time they're about six years old, 00:45:33.405 --> 00:45:35.030 align:middle line:84% for English speaking kids, the estimate 00:45:35.030 --> 00:45:36.980 align:middle line:84% is that they know about 13,000 words. 00:45:36.980 --> 00:45:39.320 align:middle line:84% So there's this period where they're learning about six 00:45:39.320 --> 00:45:42.500 align:middle line:84% or seven words every day, which is astonishing. 00:45:42.500 --> 00:45:44.600 align:middle line:84% I mean, if you imagine trying to learn 00:45:44.600 --> 00:45:47.660 align:middle line:84% six or seven words every day, go study some foreign language 00:45:47.660 --> 00:45:49.820 align:middle line:84% and try to learn six or seven words a day. 00:45:49.820 --> 00:45:51.890 align:middle line:90% That's a really fast clip. 00:45:51.890 --> 00:45:53.993 align:middle line:84% But if you can do it for six years, 00:45:53.993 --> 00:45:55.910 align:middle line:84% then you will speak that language really well, 00:45:55.910 --> 00:45:58.650 align:middle line:90% whatever it is. 00:45:58.650 --> 00:46:01.470 align:middle line:84% This has all been about phonology. 00:46:01.470 --> 00:46:04.380 align:middle line:84% There's a lot of work on the acquisition of syntax. 00:46:04.380 --> 00:46:09.120 align:middle line:84% And again, this is sort of like the Japanese child 00:46:09.120 --> 00:46:12.570 align:middle line:84% phonology that I was showing you at the very beginning here. 00:46:12.570 --> 00:46:13.810 align:middle line:90% A similar kind of case-- 00:46:13.810 --> 00:46:16.440 align:middle line:84% there are lots of cases of children doing things 00:46:16.440 --> 00:46:17.850 align:middle line:90% that adults don't do. 00:46:17.850 --> 00:46:20.310 align:middle line:84% And yet, in the course of doing that, 00:46:20.310 --> 00:46:22.230 align:middle line:84% demonstrating that they know more than you 00:46:22.230 --> 00:46:23.490 align:middle line:90% might have expected them to. 00:46:23.490 --> 00:46:26.250 align:middle line:84% So children in many languages go through what's 00:46:26.250 --> 00:46:29.040 align:middle line:90% called a root infinitive stage. 00:46:29.040 --> 00:46:34.500 align:middle line:84% That is, they use verb forms that are infinitives-- 00:46:34.500 --> 00:46:38.968 align:middle line:84% the kind of verb that you would use in an infinitival clause. 00:46:38.968 --> 00:46:41.010 align:middle line:84% Here are some examples from Danish and from Dutch 00:46:41.010 --> 00:46:42.180 align:middle line:90% and from French. 00:46:42.180 --> 00:46:43.268 align:middle line:90% This is not universal. 00:46:43.268 --> 00:46:45.810 align:middle line:84% There are languages that don't use optional root infinitives. 00:46:45.810 --> 00:46:47.490 align:middle line:90% But many languages do. 00:46:47.490 --> 00:46:49.380 align:middle line:90% English, arguably, has it. 00:46:49.380 --> 00:46:51.210 align:middle line:84% Of course, English verb morphology 00:46:51.210 --> 00:46:54.090 align:middle line:84% is so impoverished that it's hard to see. 00:46:54.090 --> 00:47:02.640 align:middle line:84% But the idea is that when children say things like, what? 00:47:02.640 --> 00:47:06.150 align:middle line:84% "Doll sleep," which is the kind of thing 00:47:06.150 --> 00:47:11.190 align:middle line:84% that a two-year-old might say, that that's a way for the child 00:47:11.190 --> 00:47:13.020 align:middle line:90% to say "The doll sleeps." 00:47:13.020 --> 00:47:14.400 align:middle line:90% And they're using an infinitive. 00:47:14.400 --> 00:47:16.650 align:middle line:84% They're using an optional infinitive, so a verb that's 00:47:16.650 --> 00:47:18.830 align:middle line:84% missing the morphology that you would expect it 00:47:18.830 --> 00:47:21.690 align:middle line:90% to have if they were an adult. 00:47:21.690 --> 00:47:25.140 align:middle line:84% But the observation is that although they are confused 00:47:25.140 --> 00:47:28.320 align:middle line:84% about whether it's OK to have a main clause be 00:47:28.320 --> 00:47:30.180 align:middle line:84% an infinitive, what's interesting 00:47:30.180 --> 00:47:32.820 align:middle line:84% is that they know all kinds of sophisticated stuff 00:47:32.820 --> 00:47:35.200 align:middle line:84% about what it would mean for it to be an infinitive. 00:47:35.200 --> 00:47:37.920 align:middle line:84% So for example, when we were talking about German, 00:47:37.920 --> 00:47:40.600 align:middle line:84% I think I told you German's a V2 language. 00:47:40.600 --> 00:47:44.490 align:middle line:84% So in ordinary main clauses, the verb is in second position. 00:47:44.490 --> 00:47:46.950 align:middle line:84% In infinitives-- well, infinitives are usually not 00:47:46.950 --> 00:47:48.480 align:middle line:90% main clauses in German. 00:47:48.480 --> 00:47:51.535 align:middle line:84% So the verb is typically not in second position. 00:47:51.535 --> 00:47:52.785 align:middle line:90% It's at the end of the clause. 00:47:52.785 --> 00:47:54.360 align:middle line:84% It goes at the end of the clause. 00:47:54.360 --> 00:47:57.420 align:middle line:84% And German-speaking children go through this root infinitive 00:47:57.420 --> 00:48:00.480 align:middle line:84% stage, a stage when they're around two or three years old, 00:48:00.480 --> 00:48:02.190 align:middle line:84% where they use infinitival verbs a lot. 00:48:02.190 --> 00:48:04.530 align:middle line:84% And they put those verbs where they should. 00:48:04.530 --> 00:48:06.757 align:middle line:84% So they are confused about something. 00:48:06.757 --> 00:48:09.090 align:middle line:84% They're confused about whether it's OK for a main clause 00:48:09.090 --> 00:48:10.050 align:middle line:90% to be infinitive. 00:48:10.050 --> 00:48:12.540 align:middle line:84% But they're not confused about the rules for where 00:48:12.540 --> 00:48:13.650 align:middle line:90% the verb goes in German. 00:48:13.650 --> 00:48:16.590 align:middle line:84% So when they're using-- when they do use tense verbs-- 00:48:16.590 --> 00:48:18.570 align:middle line:84% and they do-- they put them in second position. 00:48:18.570 --> 00:48:20.520 align:middle line:84% And when they use the infinitival verbs, 00:48:20.520 --> 00:48:22.320 align:middle line:90% they put them at the end. 00:48:22.320 --> 00:48:24.483 align:middle line:84% And that kind of fact has been replicated 00:48:24.483 --> 00:48:25.900 align:middle line:84% in a bunch of different languages. 00:48:25.900 --> 00:48:29.085 align:middle line:84% So here's a German two-year-old saying-- 00:48:29.085 --> 00:48:31.095 align:middle line:84% the child wants to say "Thorsten has Caesar"-- 00:48:31.095 --> 00:48:34.260 align:middle line:90% 00:48:34.260 --> 00:48:37.500 align:middle line:84% I think Thorsten was a teddy bear-- 00:48:37.500 --> 00:48:39.720 align:middle line:90% and "have" is in the infinitive. 00:48:39.720 --> 00:48:41.400 align:middle line:90% And so it's at the end. 00:48:41.400 --> 00:48:44.640 align:middle line:84% But in that second example, the child is trying to say, 00:48:44.640 --> 00:48:46.027 align:middle line:90% "I have a big ball." 00:48:46.027 --> 00:48:48.360 align:middle line:84% And that's pretty good German, apart from mispronouncing 00:48:48.360 --> 00:48:50.340 align:middle line:90% the word for "big." 00:48:50.340 --> 00:48:52.920 align:middle line:84% The word for "have" is in second position. 00:48:52.920 --> 00:48:54.120 align:middle line:90% It's inflected. 00:48:54.120 --> 00:48:55.200 align:middle line:90% It's not infinitival. 00:48:55.200 --> 00:48:58.560 align:middle line:90% And it goes in second position. 00:48:58.560 --> 00:49:01.560 align:middle line:84% So yeah, this child is acting like an adult German, 00:49:01.560 --> 00:49:05.520 align:middle line:84% except for the fact about having an infinitive in a main clause. 00:49:05.520 --> 00:49:09.000 align:middle line:84% So there's a mistake, but it's a constrained kind of mistake. 00:49:09.000 --> 00:49:09.990 align:middle line:90% And this is pretty-- 00:49:09.990 --> 00:49:12.540 align:middle line:90% 00:49:12.540 --> 00:49:15.060 align:middle line:84% again, this is-- we're back now to studying 00:49:15.060 --> 00:49:18.760 align:middle line:84% this stuff by just studying what children actually say. 00:49:18.760 --> 00:49:21.480 align:middle line:84% This is a recording study of this German child, 00:49:21.480 --> 00:49:24.540 align:middle line:84% Andreas, who was two at the time-- 00:49:24.540 --> 00:49:26.130 align:middle line:90% 281 utterances. 00:49:26.130 --> 00:49:32.520 align:middle line:84% And a healthy proportion of them were these root infinitives. 00:49:32.520 --> 00:49:35.010 align:middle line:84% And if you look at them, it turns out 00:49:35.010 --> 00:49:37.470 align:middle line:84% that the finite verbs are in the second position mostly 00:49:37.470 --> 00:49:40.770 align:middle line:84% and the infinitives are in final position mostly. 00:49:40.770 --> 00:49:43.080 align:middle line:84% So he's confused about where he can use infinitives. 00:49:43.080 --> 00:49:45.205 align:middle line:84% But he knows what to do with them once he has them. 00:49:45.205 --> 00:49:49.750 align:middle line:90% 00:49:49.750 --> 00:49:52.330 align:middle line:84% Yeah, we have time for this, I think. 00:49:52.330 --> 00:49:56.140 align:middle line:84% OK, so all of this has been about studying 00:49:56.140 --> 00:49:58.120 align:middle line:90% children's naturalistic output. 00:49:58.120 --> 00:50:01.930 align:middle line:84% That is, it's been about studying what children actually 00:50:01.930 --> 00:50:03.940 align:middle line:84% say, so making careful recordings. 00:50:03.940 --> 00:50:08.140 align:middle line:84% So a lot of this involves putting tape recorders 00:50:08.140 --> 00:50:10.480 align:middle line:84% in the homes of parents of young children. 00:50:10.480 --> 00:50:13.740 align:middle line:90% The parents volunteer for this. 00:50:13.740 --> 00:50:14.740 align:middle line:90% Hopefully, they're paid. 00:50:14.740 --> 00:50:15.670 align:middle line:84% I don't know if they're paid or not. 00:50:15.670 --> 00:50:18.190 align:middle line:84% Anyway, recording devices are put in their houses. 00:50:18.190 --> 00:50:20.140 align:middle line:84% And you end up with many, many recordings 00:50:20.140 --> 00:50:23.125 align:middle line:84% of children speaking, parents speaking to children. 00:50:23.125 --> 00:50:24.790 align:middle line:84% And you transcribe these recordings. 00:50:24.790 --> 00:50:27.700 align:middle line:84% There's an online database of a lot 00:50:27.700 --> 00:50:29.230 align:middle line:90% of these classic recordings. 00:50:29.230 --> 00:50:34.000 align:middle line:84% It's called the CHILDES database, which 00:50:34.000 --> 00:50:36.100 align:middle line:90% I think is publicly accessible. 00:50:36.100 --> 00:50:40.310 align:middle line:84% Actually, if you Google it, you can probably find it. 00:50:40.310 --> 00:50:42.640 align:middle line:84% But there's another kind of work that people do-- 00:50:42.640 --> 00:50:45.940 align:middle line:84% and we have a lab here at MIT where this kind of work 00:50:45.940 --> 00:50:47.440 align:middle line:90% is being done-- 00:50:47.440 --> 00:50:49.540 align:middle line:84% where you do what you could think of as consultant 00:50:49.540 --> 00:50:54.178 align:middle line:84% work, kind of the work you're doing with the languages 00:50:54.178 --> 00:50:55.720 align:middle line:84% that you've been doing field work on, 00:50:55.720 --> 00:50:59.680 align:middle line:84% except the children are probably more 00:50:59.680 --> 00:51:05.390 align:middle line:84% likely to cry and to demand cookies and things like that. 00:51:05.390 --> 00:51:08.530 align:middle line:84% So some of this consultant work involves 00:51:08.530 --> 00:51:11.170 align:middle line:90% showing a child a story-- 00:51:11.170 --> 00:51:13.090 align:middle line:84% either telling them a story or showing them 00:51:13.090 --> 00:51:15.190 align:middle line:84% a series of pictures that demonstrate the story-- 00:51:15.190 --> 00:51:20.150 align:middle line:84% and then asking them whether something is true or not. 00:51:20.150 --> 00:51:24.760 align:middle line:84% So one classic way to do this kind of experiment-- you 00:51:24.760 --> 00:51:27.760 align:middle line:84% have somebody who is telling the story, 00:51:27.760 --> 00:51:30.160 align:middle line:84% and then another experimenter who has a puppet. 00:51:30.160 --> 00:51:32.920 align:middle line:84% And the child's job is to tell the puppet whether the puppet 00:51:32.920 --> 00:51:34.852 align:middle line:90% is right or wrong. 00:51:34.852 --> 00:51:36.310 align:middle line:84% The people who do this kind of work 00:51:36.310 --> 00:51:39.280 align:middle line:84% have discovered it's easier to get a child to tell a puppet 00:51:39.280 --> 00:51:42.100 align:middle line:84% that the puppet is wrong than to tell an adult that the adult is 00:51:42.100 --> 00:51:42.970 align:middle line:90% wrong. 00:51:42.970 --> 00:51:45.700 align:middle line:84% So you have the puppet say, I know what happened. 00:51:45.700 --> 00:51:47.800 align:middle line:84% And then the puppet describes what happened. 00:51:47.800 --> 00:51:51.310 align:middle line:84% And that way, you learn whether the child 00:51:51.310 --> 00:51:54.400 align:middle line:84% thinks that the sentence that the puppet utters 00:51:54.400 --> 00:51:57.920 align:middle line:84% is a true description of what has happened or not. 00:51:57.920 --> 00:52:00.380 align:middle line:84% So there's a lot of work, for example, 00:52:00.380 --> 00:52:03.160 align:middle line:84% on something that children do, which is kind of interesting. 00:52:03.160 --> 00:52:09.100 align:middle line:84% But if you show children a picture-- 00:52:09.100 --> 00:52:11.350 align:middle line:84% I've done this work with five-year-olds. 00:52:11.350 --> 00:52:14.020 align:middle line:90% This goes on fairly late. 00:52:14.020 --> 00:52:21.910 align:middle line:84% If you show children a picture of, let's say, four elephants-- 00:52:21.910 --> 00:52:24.590 align:middle line:84% and now, you will have to apologize-- 00:52:24.590 --> 00:52:29.170 align:middle line:84% you will have to forgive my inability to draw an elephant. 00:52:29.170 --> 00:52:31.600 align:middle line:84% Please pretend that these are elephants. 00:52:31.600 --> 00:52:34.028 align:middle line:84% Here, let's make it a story about three elephants. 00:52:34.028 --> 00:52:36.070 align:middle line:84% So there are three elephants, you tell the child. 00:52:36.070 --> 00:52:40.480 align:middle line:90% And you tell them there are-- 00:52:40.480 --> 00:52:42.490 align:middle line:90% and there are rabbits sitting-- 00:52:42.490 --> 00:52:44.230 align:middle line:90% here some rabbits. 00:52:44.230 --> 00:52:46.720 align:middle line:84% Why am I telling the story this way? 00:52:46.720 --> 00:52:50.724 align:middle line:84% Why did I not put these pictures on the slide? 00:52:50.724 --> 00:52:53.225 align:middle line:84% OK, and there are rabbits on two of the elephants. 00:52:53.225 --> 00:52:53.725 align:middle line:90% Yes? 00:52:53.725 --> 00:52:56.410 align:middle line:84% AUDIENCE: Those are remarkably good elephants and rabbits. 00:52:56.410 --> 00:52:57.550 align:middle line:84% NORVIN RICHARDS: Oh thank you very much. 00:52:57.550 --> 00:52:58.150 align:middle line:90% Thank you. 00:52:58.150 --> 00:52:59.530 align:middle line:84% After all that self-deprecation, that's 00:52:59.530 --> 00:53:01.030 align:middle line:84% exactly what I was hoping somebody would say. 00:53:01.030 --> 00:53:02.822 align:middle line:84% Please write that on the course evaluation: 00:53:02.822 --> 00:53:05.630 align:middle line:84% "Surprisingly good at drawing elephants and rabbits." 00:53:05.630 --> 00:53:07.510 align:middle line:84% So here are three elephants and two rabbits. 00:53:07.510 --> 00:53:09.823 align:middle line:84% You show a child a picture like this that has 00:53:09.823 --> 00:53:10.990 align:middle line:90% elephants and rabbits in it. 00:53:10.990 --> 00:53:16.060 align:middle line:84% And you ask-- the puppet says, "I know what's happening. 00:53:16.060 --> 00:53:20.275 align:middle line:84% Every rabbit is riding an elephant." 00:53:20.275 --> 00:53:25.310 align:middle line:90% 00:53:25.310 --> 00:53:28.010 align:middle line:84% And an adult-- you also do this experiment with adults, just 00:53:28.010 --> 00:53:28.760 align:middle line:90% to have a control. 00:53:28.760 --> 00:53:31.135 align:middle line:84% So you tell the adults, this is going to be for children, 00:53:31.135 --> 00:53:32.083 align:middle line:90% but please humor us. 00:53:32.083 --> 00:53:32.750 align:middle line:90% Here's a puppet. 00:53:32.750 --> 00:53:34.790 align:middle line:84% You have to tell the puppet whether it's right or wrong. 00:53:34.790 --> 00:53:36.060 align:middle line:84% The puppet looks at this picture and says, 00:53:36.060 --> 00:53:36.977 align:middle line:90% "I know what happened. 00:53:36.977 --> 00:53:38.750 align:middle line:84% Every rabbit is riding an elephant." 00:53:38.750 --> 00:53:41.480 align:middle line:90% And adults will say yes. 00:53:41.480 --> 00:53:44.090 align:middle line:84% But children at a certain age will say no. 00:53:44.090 --> 00:53:46.940 align:middle line:84% And then when you ask them why not, they'll 00:53:46.940 --> 00:53:49.490 align:middle line:84% point at this elephant and say, "This elephant 00:53:49.490 --> 00:53:51.687 align:middle line:90% doesn't have a rabbit." 00:53:51.687 --> 00:53:53.270 align:middle line:84% And so people are trying to figure out 00:53:53.270 --> 00:53:55.910 align:middle line:84% what is going on in the heads of these children. 00:53:55.910 --> 00:53:57.995 align:middle line:90% Why are they doing this? 00:53:57.995 --> 00:53:59.870 align:middle line:84% So this is an active area of research, trying 00:53:59.870 --> 00:54:00.620 align:middle line:90% to figure out what's going on. 00:54:00.620 --> 00:54:01.120 align:middle line:90% Yeah? 00:54:01.120 --> 00:54:04.318 align:middle line:84% AUDIENCE: Isn't that complicated by syntax [INAUDIBLE] 00:54:04.318 --> 00:54:06.110 align:middle line:84% because when I first heard that [INAUDIBLE] 00:54:06.110 --> 00:54:10.243 align:middle line:90% was a nightmare [INAUDIBLE] 00:54:10.243 --> 00:54:11.660 align:middle line:84% NORVIN RICHARDS: Oh yeah, you were 00:54:11.660 --> 00:54:12.890 align:middle line:84% traumatized by the guards standing 00:54:12.890 --> 00:54:13.932 align:middle line:90% in front of the building. 00:54:13.932 --> 00:54:15.350 align:middle line:90% I'm so sorry. 00:54:15.350 --> 00:54:16.910 align:middle line:84% Every rabbit is riding an elephant. 00:54:16.910 --> 00:54:21.650 align:middle line:84% Yeah, so the thing is the children-- 00:54:21.650 --> 00:54:22.840 align:middle line:90% how shall I say this? 00:54:22.840 --> 00:54:24.890 align:middle line:84% I mean, so first of all, if you have the puppet 00:54:24.890 --> 00:54:27.872 align:middle line:84% say this, the children-- so the children will say, no, 00:54:27.872 --> 00:54:28.580 align:middle line:90% that's not right. 00:54:28.580 --> 00:54:31.400 align:middle line:84% And when you ask them why not, they point at the elephant. 00:54:31.400 --> 00:54:32.900 align:middle line:90% They don't panic. 00:54:32.900 --> 00:54:34.780 align:middle line:90% They don't like start to cry. 00:54:34.780 --> 00:54:35.780 align:middle line:90% I mean, some of them do. 00:54:35.780 --> 00:54:36.810 align:middle line:90% You exclude those. 00:54:36.810 --> 00:54:38.660 align:middle line:84% But the ones who can handle this experiment, 00:54:38.660 --> 00:54:44.810 align:middle line:84% they have a theory about what the sentence means, apparently. 00:54:44.810 --> 00:54:48.800 align:middle line:84% And then if you give children another picture which is just 00:54:48.800 --> 00:54:51.220 align:middle line:84% like this one but there's an extra rabbit, 00:54:51.220 --> 00:54:54.410 align:middle line:84% then the adults-- there's the third rabbit. 00:54:54.410 --> 00:54:57.500 align:middle line:84% Then the adult-- that's about as large as an elephant. 00:54:57.500 --> 00:54:59.450 align:middle line:84% Then the adults and the children agree. 00:54:59.450 --> 00:55:01.160 align:middle line:84% They all say, oh no, now it's not true. 00:55:01.160 --> 00:55:03.410 align:middle line:84% It's not true that every rabbit is riding an elephant. 00:55:03.410 --> 00:55:07.220 align:middle line:84% So children apparently have some-- at a certain age, 00:55:07.220 --> 00:55:08.990 align:middle line:84% have something interesting that they're 00:55:08.990 --> 00:55:11.930 align:middle line:84% doing with qualifiers that is not what adults are doing. 00:55:11.930 --> 00:55:12.910 align:middle line:90% Yeah? 00:55:12.910 --> 00:55:15.320 align:middle line:84% AUDIENCE: In this picture with the rabbit, 00:55:15.320 --> 00:55:18.770 align:middle line:84% do the children point to the rabbit [INAUDIBLE] not true 00:55:18.770 --> 00:55:19.340 align:middle line:90% always? 00:55:19.340 --> 00:55:20.017 align:middle line:90% [INAUDIBLE] 00:55:20.017 --> 00:55:22.100 align:middle line:84% NORVIN RICHARDS: Or do they point to the elephant? 00:55:22.100 --> 00:55:22.760 align:middle line:90% Oh, I'm sorry. 00:55:22.760 --> 00:55:24.552 align:middle line:84% Yeah, I should also have said, you can also 00:55:24.552 --> 00:55:27.290 align:middle line:84% point-- you should give them a picture like this, where 00:55:27.290 --> 00:55:31.970 align:middle line:84% every elephant has-- where there isn't an extra elephant, 00:55:31.970 --> 00:55:33.260 align:middle line:90% but there is an extra rabbit. 00:55:33.260 --> 00:55:35.300 align:middle line:84% And then again, they will say, "No, that's not true." 00:55:35.300 --> 00:55:36.200 align:middle line:90% And they'll point to the rabbit. 00:55:36.200 --> 00:55:37.550 align:middle line:84% I actually don't know if you give them 00:55:37.550 --> 00:55:39.650 align:middle line:84% an extra of both of them whether they always point 00:55:39.650 --> 00:55:40.850 align:middle line:90% the elephant or the rabbit. 00:55:40.850 --> 00:55:44.582 align:middle line:84% That's a good question, interesting to find out. 00:55:44.582 --> 00:55:46.540 align:middle line:84% So this is the kind of thing that people study. 00:55:46.540 --> 00:55:49.050 align:middle line:84% They do all this interesting work with children, 00:55:49.050 --> 00:55:50.800 align:middle line:84% try to figure out what exactly is going on 00:55:50.800 --> 00:55:52.330 align:middle line:90% in their little heads. 00:55:52.330 --> 00:55:56.650 align:middle line:84% And one of the big takeaways from this research 00:55:56.650 --> 00:55:58.150 align:middle line:84% is that they-- first of all, they're 00:55:58.150 --> 00:56:01.990 align:middle line:84% not like adults, which any parent could have told them. 00:56:01.990 --> 00:56:04.300 align:middle line:84% But also, although they are not like adults, 00:56:04.300 --> 00:56:08.410 align:middle line:84% they have mistakes of certain kinds and other kinds of things 00:56:08.410 --> 00:56:09.910 align:middle line:84% that they don't make mistakes about. 00:56:09.910 --> 00:56:11.860 align:middle line:84% So they may use optional infinitives, 00:56:11.860 --> 00:56:15.417 align:middle line:84% if they're learning a language that has optional infinitives. 00:56:15.417 --> 00:56:17.500 align:middle line:84% But they put the infinitives where the infinitives 00:56:17.500 --> 00:56:18.667 align:middle line:90% ought to go in the sentence. 00:56:18.667 --> 00:56:23.380 align:middle line:84% They've got a particular thing that they're wrong about. 00:56:23.380 --> 00:56:25.570 align:middle line:84% And it's for a particular stage of their life. 00:56:25.570 --> 00:56:26.920 align:middle line:90% They eventually age out of it. 00:56:26.920 --> 00:56:29.920 align:middle line:84% Or similarly here, this is a very common phenomenon, 00:56:29.920 --> 00:56:34.360 align:middle line:84% that children, if you show them this kind of picture, 00:56:34.360 --> 00:56:38.980 align:middle line:84% will say "No, no" because of the extra elephant. 00:56:38.980 --> 00:56:41.200 align:middle line:84% And we're still trying to figure out 00:56:41.200 --> 00:56:44.180 align:middle line:84% what exactly it is that children are confused about there. 00:56:44.180 --> 00:56:46.270 align:middle line:90% So there's lots of work to do. 00:56:46.270 --> 00:56:48.190 align:middle line:84% If you are interested in linguistics 00:56:48.190 --> 00:56:49.782 align:middle line:84% and interested in children, as I say, 00:56:49.782 --> 00:56:51.740 align:middle line:84% there's a lab that works on this kind of thing. 00:56:51.740 --> 00:56:53.590 align:middle line:84% And they're always looking for volunteers, 00:56:53.590 --> 00:56:55.300 align:middle line:90% people who want to help out. 00:56:55.300 --> 00:56:57.100 align:middle line:84% And there is a class specifically 00:56:57.100 --> 00:56:59.830 align:middle line:84% about language acquisition, undergrad level class 00:56:59.830 --> 00:57:02.860 align:middle line:84% for language acquisition, where people study 00:57:02.860 --> 00:57:05.290 align:middle line:90% how to do this kind of thing. 00:57:05.290 --> 00:57:07.823 align:middle line:90% Questions about any of this? 00:57:07.823 --> 00:57:09.240 align:middle line:84% When I put up the slides, I'll try 00:57:09.240 --> 00:57:12.950 align:middle line:84% to find the actual pictures with the elephants and the rabbits. 00:57:12.950 --> 00:57:16.580 align:middle line:84% OK, all right, well then, let's end a little bit early today. 00:57:16.580 --> 00:57:20.700 align:middle line:84% Next week, we will start up on Tuesday. 00:57:20.700 --> 00:57:21.680 align:middle line:90% It'll be our last day. 00:57:21.680 --> 00:57:24.170 align:middle line:84% I'm debating-- oh, actually, maybe I can get feedback 00:57:24.170 --> 00:57:25.250 align:middle line:90% from you guys about this. 00:57:25.250 --> 00:57:28.100 align:middle line:84% On our last day, there are two things that I could do. 00:57:28.100 --> 00:57:31.730 align:middle line:84% I could try to do a summary of everything that we have done 00:57:31.730 --> 00:57:35.270 align:middle line:84% and, well, probably a brief summary in which we wouldn't 00:57:35.270 --> 00:57:37.970 align:middle line:84% talk for very long, and encourage you again 00:57:37.970 --> 00:57:41.870 align:middle line:84% to do the evaluation forms, and then send you on your way. 00:57:41.870 --> 00:57:46.820 align:middle line:84% Or I could do a class on signed language, 00:57:46.820 --> 00:57:49.650 align:middle line:84% which is a topic that I haven't gotten to during the semester. 00:57:49.650 --> 00:57:53.040 align:middle line:84% But we could spend a spend a class doing that. 00:57:53.040 --> 00:57:54.870 align:middle line:84% Think about that and send me an email. 00:57:54.870 --> 00:57:56.750 align:middle line:84% Some of you are using your facial expressions 00:57:56.750 --> 00:57:59.167 align:middle line:84% to give me ideas about which of those things you'd prefer. 00:57:59.167 --> 00:58:01.565 align:middle line:84% But send me email if you have a preference one way 00:58:01.565 --> 00:58:03.690 align:middle line:84% or another about which of those things I should do. 00:58:03.690 --> 00:58:06.740 align:middle line:84% And I will take your preferences into account. 00:58:06.740 --> 00:58:10.630 align:middle line:84% All right, have a good rest of your day. 00:58:10.630 --> 00:58:15.000 align:middle line:90%