WEBVTT 00:00:00.000 --> 00:00:02.405 align:middle line:90% [SQUEAKING] 00:00:02.405 --> 00:00:04.329 align:middle line:90% [RUSTLING] 00:00:04.329 --> 00:00:06.734 align:middle line:90% [CLICKING] 00:00:06.734 --> 00:00:10.590 align:middle line:90% 00:00:10.590 --> 00:00:12.420 align:middle line:90% NORVIN RICHARDS: So morphology. 00:00:12.420 --> 00:00:14.700 align:middle line:90% Remember morphology? 00:00:14.700 --> 00:00:17.700 align:middle line:84% We were talking when we talked about morphology about the fact 00:00:17.700 --> 00:00:21.120 align:middle line:84% that a word like "unlockable" is ambiguous. 00:00:21.120 --> 00:00:25.290 align:middle line:84% It can mean either it's possible to unlock it, 00:00:25.290 --> 00:00:29.130 align:middle line:84% or it is not possible to lock it. 00:00:29.130 --> 00:00:34.285 align:middle line:84% So it can be a desirable property of a lock or not. 00:00:34.285 --> 00:00:35.910 align:middle line:84% When we talked, I think, about the fact 00:00:35.910 --> 00:00:38.195 align:middle line:84% that at least I want to pronounce 00:00:38.195 --> 00:00:39.570 align:middle line:84% it a little differently depending 00:00:39.570 --> 00:00:41.670 align:middle line:90% on which of those things I say. 00:00:41.670 --> 00:00:48.670 align:middle line:84% So The door is unlockable" means it can be unlocked. 00:00:48.670 --> 00:00:52.180 align:middle line:84% But the door is broken, "It's unlockable," 00:00:52.180 --> 00:00:53.700 align:middle line:90% means it cannot be locked. 00:00:53.700 --> 00:00:57.600 align:middle line:84% I have this desire to put an extra oomph on "un-"-- 00:00:57.600 --> 00:01:00.440 align:middle line:84% an extra little demi-stress beat on "un-" 00:01:00.440 --> 00:01:01.815 align:middle line:84% if I mean the thing on the right. 00:01:01.815 --> 00:01:04.709 align:middle line:90% 00:01:04.709 --> 00:01:06.960 align:middle line:84% And what we said, when we were talking about this, 00:01:06.960 --> 00:01:10.345 align:middle line:84% was we can account for this kind of ambiguity 00:01:10.345 --> 00:01:11.220 align:middle line:90% in the following way. 00:01:11.220 --> 00:01:14.130 align:middle line:84% We'll say some things which are clearly true. 00:01:14.130 --> 00:01:18.025 align:middle line:84% First, "-able" combines with verbs to make adjectives. 00:01:18.025 --> 00:01:19.650 align:middle line:84% There are a bunch of verbs that you can 00:01:19.650 --> 00:01:20.940 align:middle line:90% make into adjectives this way. 00:01:20.940 --> 00:01:22.815 align:middle line:84% So you can take a verb like "sing," or a verb 00:01:22.815 --> 00:01:24.720 align:middle line:84% like "understand" and get adjectives 00:01:24.720 --> 00:01:26.910 align:middle line:84% like "singable" or "understandable." 00:01:26.910 --> 00:01:28.560 align:middle line:84% And to say that something is singable 00:01:28.560 --> 00:01:30.830 align:middle line:84% is to say that it's possible to sing it, right? 00:01:30.830 --> 00:01:32.100 align:middle line:90% That's what that means. 00:01:32.100 --> 00:01:36.000 align:middle line:84% "Understandable" means it's possible to understand it. 00:01:36.000 --> 00:01:41.040 align:middle line:84% So there's a "-able" suffix that changes verbs into adjectives. 00:01:41.040 --> 00:01:43.850 align:middle line:84% And then, we said, there are two "un-"s. 00:01:43.850 --> 00:01:48.280 align:middle line:84% There's an "un-" that combines with verbs and makes verbs. 00:01:48.280 --> 00:01:53.370 align:middle line:84% That's an "un-" that means something like undo the effects 00:01:53.370 --> 00:01:57.960 align:middle line:84% of, or change something so that it is no longer in the state 00:01:57.960 --> 00:02:01.110 align:middle line:84% that it would have been if the verb had applied to it, 00:02:01.110 --> 00:02:02.160 align:middle line:90% or something like that. 00:02:02.160 --> 00:02:06.780 align:middle line:84% So "untie" means "take something and do things to it such 00:02:06.780 --> 00:02:08.669 align:middle line:84% that it is no longer in the state 00:02:08.669 --> 00:02:11.490 align:middle line:84% that it is standardly in after you have tied it." 00:02:11.490 --> 00:02:13.920 align:middle line:84% Or to put that a little more briefly, "take something 00:02:13.920 --> 00:02:16.860 align:middle line:84% that was tied and make it so it is not tied." 00:02:16.860 --> 00:02:19.060 align:middle line:90% That's what "untie" means. 00:02:19.060 --> 00:02:21.250 align:middle line:84% Similarly for "undo" and a bunch of other "un-"s. 00:02:21.250 --> 00:02:21.750 align:middle line:90% Yep. 00:02:21.750 --> 00:02:23.967 align:middle line:84% So there's an "un-" that applies to verbs, and has-- 00:02:23.967 --> 00:02:25.800 align:middle line:84% it's sometimes called a reversitive meaning. 00:02:25.800 --> 00:02:28.060 align:middle line:84% You change it so that it's back from the state 00:02:28.060 --> 00:02:30.810 align:middle line:90% it would have been in. 00:02:30.810 --> 00:02:33.240 align:middle line:84% And then there's another "un-," let us call it "un-" 00:02:33.240 --> 00:02:34.500 align:middle line:90% number two. 00:02:34.500 --> 00:02:37.740 align:middle line:84% "Un-" number two combines with adjectives and makes adjectives 00:02:37.740 --> 00:02:42.210 align:middle line:84% that mean more or less "not (adjective)." 00:02:42.210 --> 00:02:46.200 align:middle line:84% So "unkind," or "unfamiliar," "unfortunate." 00:02:46.200 --> 00:02:49.840 align:middle line:84% These all mean not the adjective, whatever it is. 00:02:49.840 --> 00:02:51.420 align:middle line:84% And so what we said was-- oh, hey. 00:02:51.420 --> 00:02:52.017 align:middle line:90% Raquel? 00:02:52.017 --> 00:02:54.350 align:middle line:84% AUDIENCE: So I have a horrible thought, and it's random. 00:02:54.350 --> 00:02:55.536 align:middle line:90% NORVIN RICHARDS: Oh, no. 00:02:55.536 --> 00:02:57.840 align:middle line:84% AUDIENCE: But when you were saying, put something so 00:02:57.840 --> 00:02:59.550 align:middle line:84% that it's no longer tied, I would 00:02:59.550 --> 00:03:02.580 align:middle line:84% argue that not only are we saying make it 00:03:02.580 --> 00:03:05.130 align:middle line:84% so it's no longer tied, but leave it in a similar state 00:03:05.130 --> 00:03:07.560 align:middle line:84% that it was at the beginning chronologically, 00:03:07.560 --> 00:03:10.110 align:middle line:84% because you can chop a shoelace in lots of little pieces, 00:03:10.110 --> 00:03:12.600 align:middle line:90% and it's no longer tied. 00:03:12.600 --> 00:03:14.363 align:middle line:90% [INAUDIBLE] 00:03:14.363 --> 00:03:16.280 align:middle line:84% NORVIN RICHARDS: Oh, well that's a nice point. 00:03:16.280 --> 00:03:16.910 align:middle line:90% Yeah. 00:03:16.910 --> 00:03:20.390 align:middle line:90% So if I take-- 00:03:20.390 --> 00:03:21.020 align:middle line:90% let's see. 00:03:21.020 --> 00:03:24.920 align:middle line:84% If I untie a shoelace, first of all, it has to start off tied, 00:03:24.920 --> 00:03:25.880 align:middle line:90% is that right? 00:03:25.880 --> 00:03:30.350 align:middle line:84% So if I have a shoelace which is not tied, I can't untie it. 00:03:30.350 --> 00:03:32.120 align:middle line:90% Yeah. 00:03:32.120 --> 00:03:36.410 align:middle line:84% But if it's tied and then I take scissors 00:03:36.410 --> 00:03:40.130 align:middle line:84% and I cut it into many small pieces, have I untied it? 00:03:40.130 --> 00:03:41.600 align:middle line:90% No, surely not. 00:03:41.600 --> 00:03:44.660 align:middle line:90% 00:03:44.660 --> 00:03:46.610 align:middle line:84% When Alexander cut the Gordian knot, 00:03:46.610 --> 00:03:48.620 align:middle line:90% he wasn't untying the knot. 00:03:48.620 --> 00:03:52.400 align:middle line:84% He was being more direct than that. 00:03:52.400 --> 00:03:54.763 align:middle line:90% Good point. 00:03:54.763 --> 00:03:56.680 align:middle line:84% So you have to put it-- so what you just said, 00:03:56.680 --> 00:03:58.030 align:middle line:90% I like the way you just said it. 00:03:58.030 --> 00:03:59.530 align:middle line:84% You have to put it back in the state 00:03:59.530 --> 00:04:02.467 align:middle line:84% that it was in before it was tied. 00:04:02.467 --> 00:04:03.550 align:middle line:90% Is that the way to say it? 00:04:03.550 --> 00:04:05.230 align:middle line:84% AUDIENCE: Going back and tie it, [? essentially. ?] 00:04:05.230 --> 00:04:06.147 align:middle line:90% NORVIN RICHARDS: Yeah. 00:04:06.147 --> 00:04:06.880 align:middle line:90% Yeah. 00:04:06.880 --> 00:04:10.870 align:middle line:90% So yeah, that's a nice point. 00:04:10.870 --> 00:04:12.018 align:middle line:90% Undo. 00:04:12.018 --> 00:04:14.560 align:middle line:84% I mean, because what we said-- we talked about this in class, 00:04:14.560 --> 00:04:18.640 align:middle line:90% that there are lots of things. 00:04:18.640 --> 00:04:20.019 align:middle line:90% That it's hard to do this. 00:04:20.019 --> 00:04:22.600 align:middle line:84% This "un-" is kind of picky about what it can combine with. 00:04:22.600 --> 00:04:25.270 align:middle line:84% So the "un-" that attaches to adjectives can attach to lots 00:04:25.270 --> 00:04:27.880 align:middle line:90% of adjectives, but you can't-- 00:04:27.880 --> 00:04:32.680 align:middle line:84% if I take some shoes, I can't unwash shoes, or unwash socks. 00:04:32.680 --> 00:04:34.810 align:middle line:84% That doesn't mean "take the socks 00:04:34.810 --> 00:04:37.850 align:middle line:90% and make them dirty again." 00:04:37.850 --> 00:04:41.675 align:middle line:84% You could imagine that it would, but that's not what it means. 00:04:41.675 --> 00:04:43.550 align:middle line:84% And maybe that's related to your observation. 00:04:43.550 --> 00:04:46.310 align:middle line:84% Yeah, it's interesting to think about. 00:04:46.310 --> 00:04:48.100 align:middle line:90% Joseph, did you have a-- 00:04:48.100 --> 00:04:50.236 align:middle line:90% AUDIENCE: Yeah, I was going to-- 00:04:50.236 --> 00:04:55.500 align:middle line:84% based on what Raquel said, does the final-- after you "un-" 00:04:55.500 --> 00:04:58.860 align:middle line:84% something, is that going to be able to be redone? 00:04:58.860 --> 00:05:02.415 align:middle line:84% So if I untie a shoelace by cutting it up, now-- 00:05:02.415 --> 00:05:03.780 align:middle line:90% NORVIN RICHARDS: Yeah. 00:05:03.780 --> 00:05:06.870 align:middle line:90% It can't be tied again. 00:05:06.870 --> 00:05:09.390 align:middle line:90% Well, let's see. 00:05:09.390 --> 00:05:13.650 align:middle line:84% If you undo an operation on a computer, 00:05:13.650 --> 00:05:15.900 align:middle line:84% does it have to be possible to do the operation again? 00:05:15.900 --> 00:05:17.980 align:middle line:90% I don't know, maybe. 00:05:17.980 --> 00:05:19.570 align:middle line:84% Yeah, there is a redo, isn't there? 00:05:19.570 --> 00:05:20.800 align:middle line:90% Yeah? 00:05:20.800 --> 00:05:22.123 align:middle line:90% Yeah, I wonder. 00:05:22.123 --> 00:05:23.290 align:middle line:90% Maybe that is what it means. 00:05:23.290 --> 00:05:27.160 align:middle line:90% Maybe it means-- yeah. 00:05:27.160 --> 00:05:29.680 align:middle line:84% And as you say, this is related to Raquel's point, 00:05:29.680 --> 00:05:31.180 align:middle line:84% you have to put it in a state such 00:05:31.180 --> 00:05:33.662 align:middle line:84% that the verb could apply to it again, maybe. 00:05:33.662 --> 00:05:35.080 align:middle line:90% Maybe. 00:05:35.080 --> 00:05:36.610 align:middle line:90% Yeah, so there's a lot. 00:05:36.610 --> 00:05:39.910 align:middle line:84% What I actually did on the slide was just to say, "un-" 00:05:39.910 --> 00:05:41.800 align:middle line:84% number one combines with verbs to make verbs. 00:05:41.800 --> 00:05:43.870 align:middle line:84% And then my mental notes to myself 00:05:43.870 --> 00:05:46.330 align:middle line:84% say, vamp something about what it means. 00:05:46.330 --> 00:05:48.310 align:middle line:84% You guys are doing some sophisticated thinking 00:05:48.310 --> 00:05:49.480 align:middle line:90% about what it means. 00:05:49.480 --> 00:05:51.920 align:middle line:84% It's a little complicated, figuring out what it means, 00:05:51.920 --> 00:05:52.780 align:middle line:90% as you can see. 00:05:52.780 --> 00:05:55.030 align:middle line:90% Nice point. 00:05:55.030 --> 00:05:58.330 align:middle line:84% Still, what I've said on the slide, apart from the vamping, 00:05:58.330 --> 00:05:59.120 align:middle line:90% I think is true. 00:05:59.120 --> 00:06:01.370 align:middle line:84% There's an "-able" that changes verbs into adjectives, 00:06:01.370 --> 00:06:03.520 align:middle line:84% and there are "un-"s, one that combines with verbs, 00:06:03.520 --> 00:06:05.270 align:middle line:84% and another that combines with adjectives. 00:06:05.270 --> 00:06:07.390 align:middle line:90% Does that sound right so far? 00:06:07.390 --> 00:06:12.400 align:middle line:84% And what all of this means is the ambiguity of "unlockable," 00:06:12.400 --> 00:06:15.190 align:middle line:84% we get to attribute it to basically the fact that there 00:06:15.190 --> 00:06:17.860 align:middle line:90% are two "un-"s. 00:06:17.860 --> 00:06:19.990 align:middle line:84% So you could have attached the "un-" 00:06:19.990 --> 00:06:22.480 align:middle line:84% before you attached the "-able," or you could have attached 00:06:22.480 --> 00:06:24.250 align:middle line:84% the "un-" after you attached the "-able." 00:06:24.250 --> 00:06:26.470 align:middle line:84% Because the "-able" is going to change a verb 00:06:26.470 --> 00:06:29.380 align:middle line:84% into an adjective, and "un-" can combine with either a verb 00:06:29.380 --> 00:06:30.250 align:middle line:90% or an adjective. 00:06:30.250 --> 00:06:31.750 align:middle line:84% So that was the way we were talking. 00:06:31.750 --> 00:06:38.650 align:middle line:84% So the idea was you can start by attaching "un-" on number one, 00:06:38.650 --> 00:06:44.052 align:middle line:84% so "lock," giving you this new verb, "unlock," 00:06:44.052 --> 00:06:45.760 align:middle line:84% which is related to the meaning of "lock" 00:06:45.760 --> 00:06:48.040 align:middle line:84% in mysterious and complicated ways that we've now been 00:06:48.040 --> 00:06:49.570 align:middle line:90% talking about. 00:06:49.570 --> 00:06:52.840 align:middle line:84% And then you can attach "-able" to that and give you 00:06:52.840 --> 00:06:54.970 align:middle line:90% an adjective, "unlockable." 00:06:54.970 --> 00:06:57.610 align:middle line:84% And I drew trees like this before 00:06:57.610 --> 00:06:59.110 align:middle line:84% and said, yeah, these trees are kind 00:06:59.110 --> 00:07:01.660 align:middle line:84% of a representation of the order in which you did things. 00:07:01.660 --> 00:07:03.910 align:middle line:84% That's all they're for, is to say 00:07:03.910 --> 00:07:07.150 align:middle line:84% you started by putting together those two 00:07:07.150 --> 00:07:09.670 align:middle line:84% things at the bottom of the tree, the verb and that prefix, 00:07:09.670 --> 00:07:11.590 align:middle line:90% and you created a verb. 00:07:11.590 --> 00:07:15.790 align:middle line:84% That's what that prefix does, it takes verbs and returns verbs. 00:07:15.790 --> 00:07:19.180 align:middle line:84% And then that verb gets to attach the second thing you do. 00:07:19.180 --> 00:07:20.620 align:middle line:84% That verb attaches to the suffix, 00:07:20.620 --> 00:07:24.910 align:middle line:84% and now the suffix changes that verb into an adjective. 00:07:24.910 --> 00:07:28.030 align:middle line:84% Or you can do things in the other order. 00:07:28.030 --> 00:07:30.130 align:middle line:84% You can attach "-able" to "lock," 00:07:30.130 --> 00:07:32.740 align:middle line:84% giving you an adjective, and you can take "un-" 00:07:32.740 --> 00:07:36.550 align:middle line:84% number two and attach "un-" number two to that adjective, 00:07:36.550 --> 00:07:39.730 align:middle line:84% giving you a new adjective meaning "not lockable." 00:07:39.730 --> 00:07:41.980 align:middle line:84% And so "unlockable" is ambiguous. 00:07:41.980 --> 00:07:45.677 align:middle line:84% And the ambiguity, we said, comes from the fact that, well, 00:07:45.677 --> 00:07:48.010 align:middle line:84% there are two "un-"s, which is something we can observe. 00:07:48.010 --> 00:07:49.480 align:middle line:84% There's an "un-" that goes on verbs and an "un-" 00:07:49.480 --> 00:07:51.430 align:middle line:84% that goes on adjectives, and there's a "-able" 00:07:51.430 --> 00:07:53.800 align:middle line:84% that changes verbs to adjectives, right? 00:07:53.800 --> 00:07:54.640 align:middle line:90% Yes. 00:07:54.640 --> 00:07:57.430 align:middle line:84% And that means that there's an "un-" 00:07:57.430 --> 00:07:59.597 align:middle line:84% that can go before the "-able," and there's an "un-" 00:07:59.597 --> 00:08:02.097 align:middle line:84% that can go after the "-able," and so we get this ambiguity. 00:08:02.097 --> 00:08:04.270 align:middle line:84% And the ambiguity is what we would expect it to be. 00:08:04.270 --> 00:08:05.050 align:middle line:90% Yeah? 00:08:05.050 --> 00:08:07.477 align:middle line:90% This is all review. 00:08:07.477 --> 00:08:09.810 align:middle line:84% Having said all this, I guess what we expect is that you 00:08:09.810 --> 00:08:12.810 align:middle line:84% could have two "un-"s, I didn't put this on the slide anywhere, 00:08:12.810 --> 00:08:16.410 align:middle line:84% that it would be possible to say it is "un-unlockable," 00:08:16.410 --> 00:08:19.110 align:middle line:90% which I think is true, actually. 00:08:19.110 --> 00:08:22.030 align:middle line:90% "This door is broken. 00:08:22.030 --> 00:08:23.160 align:middle line:90% It is un-unlockable." 00:08:23.160 --> 00:08:26.520 align:middle line:84% That would mean it cannot be unlocked. 00:08:26.520 --> 00:08:28.470 align:middle line:90% I think that's true. 00:08:28.470 --> 00:08:31.020 align:middle line:84% Go ask some people whose minds have not been 00:08:31.020 --> 00:08:32.340 align:middle line:90% contaminated by linguistics. 00:08:32.340 --> 00:08:36.570 align:middle line:84% Go harass your roommates or whoever. 00:08:36.570 --> 00:08:39.370 align:middle line:84% You'll make yourself popular that way. 00:08:39.370 --> 00:08:41.370 align:middle line:84% And if you are going to try to learn from me how 00:08:41.370 --> 00:08:42.900 align:middle line:84% to make yourself popular, then boy, 00:08:42.900 --> 00:08:45.870 align:middle line:90% are you in the wrong class. 00:08:45.870 --> 00:08:48.900 align:middle line:90% OK, is this all clear? 00:08:48.900 --> 00:08:50.970 align:middle line:84% So the important part of the story 00:08:50.970 --> 00:08:54.030 align:middle line:84% is to say, yeah, "unlockable" is a word. 00:08:54.030 --> 00:08:56.430 align:middle line:84% It's got three morphemes in it, a prefix, and a root, 00:08:56.430 --> 00:08:57.720 align:middle line:90% and a suffix. 00:08:57.720 --> 00:09:00.030 align:middle line:84% But it isn't just three morphemes in a row. 00:09:00.030 --> 00:09:02.850 align:middle line:84% Those morphemes were assembled in an order. 00:09:02.850 --> 00:09:04.920 align:middle line:90% You assembled them pairwise. 00:09:04.920 --> 00:09:06.510 align:middle line:84% You first put two of them together, 00:09:06.510 --> 00:09:09.270 align:middle line:84% and then you added another one to the result 00:09:09.270 --> 00:09:13.200 align:middle line:84% of that first putting together, and that order has consequences 00:09:13.200 --> 00:09:15.510 align:middle line:90% for interpretation. 00:09:15.510 --> 00:09:18.270 align:middle line:84% So it's not false to say that unlockable 00:09:18.270 --> 00:09:21.960 align:middle line:84% consists of three morphemes, a prefix, a stem, and a suffix. 00:09:21.960 --> 00:09:23.880 align:middle line:84% But it's not a complete description 00:09:23.880 --> 00:09:25.500 align:middle line:90% of what's happened either. 00:09:25.500 --> 00:09:29.460 align:middle line:84% It is that, but it is also those trees, 00:09:29.460 --> 00:09:31.350 align:middle line:84% or those trees represent something, namely 00:09:31.350 --> 00:09:33.160 align:middle line:84% the order in which you did things. 00:09:33.160 --> 00:09:35.810 align:middle line:90% Does that make sense? 00:09:35.810 --> 00:09:37.430 align:middle line:84% Now what we're going to do now is 00:09:37.430 --> 00:09:38.930 align:middle line:84% start talking about syntax, which 00:09:38.930 --> 00:09:40.555 align:middle line:84% is the study of how words are assembled 00:09:40.555 --> 00:09:44.510 align:middle line:84% to make sentences, words, sometimes things smaller 00:09:44.510 --> 00:09:45.720 align:middle line:90% than words, as we'll see. 00:09:45.720 --> 00:09:47.720 align:middle line:84% And we're going to see that it's useful to think 00:09:47.720 --> 00:09:50.300 align:middle line:84% of sentences as being put together 00:09:50.300 --> 00:09:55.160 align:middle line:84% in a bunch of operations more or less the way "unlockable" is-- 00:09:55.160 --> 00:09:59.810 align:middle line:84% that we take pairs of words and put them together 00:09:59.810 --> 00:10:05.840 align:middle line:84% to form larger objects the way I just did for "unlockable." 00:10:05.840 --> 00:10:08.960 align:middle line:90% 00:10:08.960 --> 00:10:11.150 align:middle line:84% Just to give you an example of the kind of thing 00:10:11.150 --> 00:10:12.567 align:middle line:84% we're going to talk about, I think 00:10:12.567 --> 00:10:16.040 align:middle line:84% I talked with you about this on the first day. 00:10:16.040 --> 00:10:19.700 align:middle line:84% Just as with "unlockable," it is true, 00:10:19.700 --> 00:10:21.410 align:middle line:84% but it is not a complete description 00:10:21.410 --> 00:10:23.870 align:middle line:84% to say that is three morphemes, a prefix, and a stem, 00:10:23.870 --> 00:10:25.130 align:middle line:90% and a suffix. 00:10:25.130 --> 00:10:27.380 align:middle line:84% That's true, but it's not a complete description. 00:10:27.380 --> 00:10:29.660 align:middle line:84% A complete description involves those trees 00:10:29.660 --> 00:10:33.530 align:middle line:84% or some equivalent explanation of the order 00:10:33.530 --> 00:10:35.090 align:middle line:90% in which you did things. 00:10:35.090 --> 00:10:38.810 align:middle line:84% Similarly, these two sentences, "John walked up the stairs," 00:10:38.810 --> 00:10:41.270 align:middle line:84% and "Mary looked up the reference," it's true, 00:10:41.270 --> 00:10:42.980 align:middle line:84% but it is not a complete description 00:10:42.980 --> 00:10:44.660 align:middle line:84% to say that those are two sentences that 00:10:44.660 --> 00:10:48.230 align:middle line:84% consist of a noun, a verb, a preposition, a determiner, 00:10:48.230 --> 00:10:50.680 align:middle line:90% and a noun. 00:10:50.680 --> 00:10:53.530 align:middle line:84% Did I say anything that alarmed anybody just now? 00:10:53.530 --> 00:10:58.330 align:middle line:84% So those are two sentences that consist of five words. 00:10:58.330 --> 00:10:59.290 align:middle line:90% That's true. 00:10:59.290 --> 00:11:01.945 align:middle line:84% And we can say things about what kinds of words they are. 00:11:01.945 --> 00:11:04.360 align:middle line:84% There are nouns up there, and verbs, 00:11:04.360 --> 00:11:08.410 align:middle line:84% and prepositions like "up," and I just called it a determiner, 00:11:08.410 --> 00:11:09.010 align:middle line:90% "the." 00:11:09.010 --> 00:11:10.990 align:middle line:84% People sometimes call it the definite article. 00:11:10.990 --> 00:11:15.430 align:middle line:84% You'll hear me call it the determiner a lot in this class. 00:11:15.430 --> 00:11:18.580 align:middle line:84% It's true to say that those two sentences consist of those five 00:11:18.580 --> 00:11:19.090 align:middle line:90% words. 00:11:19.090 --> 00:11:21.485 align:middle line:84% But just as with the two versions of "unlockable," 00:11:21.485 --> 00:11:23.110 align:middle line:84% we can convince ourselves that it's not 00:11:23.110 --> 00:11:24.610 align:middle line:84% a complete description, that it's 00:11:24.610 --> 00:11:28.210 align:middle line:84% helpful to think about the order in which you assembled 00:11:28.210 --> 00:11:30.290 align:middle line:90% these things. 00:11:30.290 --> 00:11:33.790 align:middle line:84% So what I told you last time was, effectively 00:11:33.790 --> 00:11:35.980 align:middle line:84% what we're going to want to say is, 00:11:35.980 --> 00:11:38.410 align:middle line:84% there was an operation that created 00:11:38.410 --> 00:11:43.000 align:middle line:84% the substring "up the stairs" in that first sentence. 00:11:43.000 --> 00:11:45.430 align:middle line:84% That's what we call a constituent. 00:11:45.430 --> 00:11:49.810 align:middle line:84% And there is no similar operation creating a substring 00:11:49.810 --> 00:11:52.480 align:middle line:84% "up the reference" in the second one. 00:11:52.480 --> 00:11:57.070 align:middle line:84% And again, this is review, but it's review from the first day. 00:11:57.070 --> 00:12:00.280 align:middle line:84% What I convinced you, I think, I hope, I tried, 00:12:00.280 --> 00:12:03.820 align:middle line:84% was that there are various syntactic phenomena, 00:12:03.820 --> 00:12:08.650 align:middle line:84% various things you get to do with sentences that treat 00:12:08.650 --> 00:12:11.620 align:middle line:84% "up the stairs" as a single object 00:12:11.620 --> 00:12:14.253 align:middle line:84% that syntax gets to manipulate in various ways. 00:12:14.253 --> 00:12:15.670 align:middle line:84% We're going be talking about that, 00:12:15.670 --> 00:12:17.920 align:middle line:90% about what that means exactly. 00:12:17.920 --> 00:12:21.700 align:middle line:84% And the same syntactic operations 00:12:21.700 --> 00:12:24.680 align:middle line:84% don't get to treat "up the reference" as a single thing. 00:12:24.680 --> 00:12:29.080 align:middle line:84% So we said it's possible to ask questions like, "Up 00:12:29.080 --> 00:12:31.150 align:middle line:90% which stairs did John walk?" 00:12:31.150 --> 00:12:33.160 align:middle line:90% It's fairly stuffy questions. 00:12:33.160 --> 00:12:36.340 align:middle line:84% It's a strange way to ask the question, but you can say it. 00:12:36.340 --> 00:12:39.430 align:middle line:84% As opposed to, "Up which reference did Mary look?" 00:12:39.430 --> 00:12:41.890 align:middle line:90% which is gibberish. 00:12:41.890 --> 00:12:44.080 align:middle line:84% So we're going to draw a distinction. 00:12:44.080 --> 00:12:46.390 align:middle line:84% This is maybe the first time that I've 00:12:46.390 --> 00:12:48.910 align:middle line:84% shown you a case where syntacticians have 00:12:48.910 --> 00:12:51.430 align:middle line:84% to care passionately about the difference between one 00:12:51.430 --> 00:12:53.140 align:middle line:84% sentence, which is complete gibberish, 00:12:53.140 --> 00:12:55.210 align:middle line:84% and another sentence, which is not great. 00:12:55.210 --> 00:12:57.220 align:middle line:84% There's a fair amount of great syntax 00:12:57.220 --> 00:12:59.920 align:middle line:84% that's built on those kinds of distinctions. 00:12:59.920 --> 00:13:02.830 align:middle line:84% Or similarly, if I say, "John walked up the stairs," 00:13:02.830 --> 00:13:05.440 align:middle line:84% and you're surprised for some reason, you can say, 00:13:05.440 --> 00:13:06.370 align:middle line:90% "Up the stairs?" 00:13:06.370 --> 00:13:08.350 align:middle line:84% That's not a weird thing for you to say. 00:13:08.350 --> 00:13:10.810 align:middle line:84% Whereas if I say, "Mary looked up the reference," 00:13:10.810 --> 00:13:13.720 align:middle line:84% and you're surprised, no matter how surprised you are, 00:13:13.720 --> 00:13:16.120 align:middle line:84% you're not going to say, "Up the reference?" 00:13:16.120 --> 00:13:19.210 align:middle line:90% That's a weird response. 00:13:19.210 --> 00:13:22.900 align:middle line:84% So the point is, just as with "unlockable," 00:13:22.900 --> 00:13:27.670 align:middle line:84% yeah, it's three morphemes, prefix, stem, suffix. 00:13:27.670 --> 00:13:30.250 align:middle line:84% But having said that, we haven't said everything. 00:13:30.250 --> 00:13:34.180 align:middle line:84% We have to know which parts of "unlockable" are single parts. 00:13:34.180 --> 00:13:37.030 align:middle line:84% Is it unattached to "lockable," or is it "unlock" 00:13:37.030 --> 00:13:38.570 align:middle line:90% with a "-able" attached to it? 00:13:38.570 --> 00:13:42.190 align:middle line:84% Those are different adjectives with different meanings. 00:13:42.190 --> 00:13:44.770 align:middle line:84% Similarly here, it's not enough to say, yeah, 00:13:44.770 --> 00:13:47.450 align:middle line:84% we've got these five words of these types. 00:13:47.450 --> 00:13:49.570 align:middle line:84% We've got to know which of these things 00:13:49.570 --> 00:13:53.200 align:middle line:84% go together, which things are parts, single parts. 00:13:53.200 --> 00:13:55.570 align:middle line:84% And there is a part, "up the stairs," 00:13:55.570 --> 00:13:58.000 align:middle line:84% what we call a constituent, that various kinds 00:13:58.000 --> 00:13:59.650 align:middle line:84% of syntactic phenomena care about, 00:13:59.650 --> 00:14:03.790 align:middle line:84% like the syntactic phenomenon can I 00:14:03.790 --> 00:14:05.410 align:middle line:90% repeat this if I'm astonished? 00:14:05.410 --> 00:14:06.790 align:middle line:90% Yeah, that's a test-- 00:14:06.790 --> 00:14:09.760 align:middle line:84% kind of test for this property of constituenthood. 00:14:09.760 --> 00:14:11.810 align:middle line:84% So there's more to a sentence than its parts. 00:14:11.810 --> 00:14:13.450 align:middle line:84% We've got to know, in what order did 00:14:13.450 --> 00:14:17.750 align:middle line:84% you put those parts together, just like with "unlockable." 00:14:17.750 --> 00:14:19.130 align:middle line:90% Yeah? 00:14:19.130 --> 00:14:21.490 align:middle line:90% Makes sense? 00:14:21.490 --> 00:14:24.760 align:middle line:90% So we're going to do syntax. 00:14:24.760 --> 00:14:28.440 align:middle line:84% We want a theory that's going to divide sentences 00:14:28.440 --> 00:14:31.200 align:middle line:90% into three kinds. 00:14:31.200 --> 00:14:32.822 align:middle line:84% There are, on the one hand, sentences 00:14:32.822 --> 00:14:34.530 align:middle line:84% that you've heard a zillion times before, 00:14:34.530 --> 00:14:36.420 align:middle line:90% like "We're going to class." 00:14:36.420 --> 00:14:40.890 align:middle line:84% And on the other hand, sentences that you have possibly 00:14:40.890 --> 00:14:44.160 align:middle line:84% never heard anyone say, but that are fine. 00:14:44.160 --> 00:14:47.250 align:middle line:84% So if I say, "My anteater is hula dancing," 00:14:47.250 --> 00:14:50.910 align:middle line:84% you may never in your life have heard anyone say that. 00:14:50.910 --> 00:14:51.660 align:middle line:90% Maybe you have. 00:14:51.660 --> 00:14:55.110 align:middle line:84% Some of you may have had more exciting lives than I have. 00:14:55.110 --> 00:14:58.710 align:middle line:90% But it's an OK sentence. 00:14:58.710 --> 00:15:01.470 align:middle line:84% As opposed to, "We're class going to," 00:15:01.470 --> 00:15:02.810 align:middle line:90% which I've given a star there. 00:15:02.810 --> 00:15:04.920 align:middle line:84% Recall that the star is what syntacticians 00:15:04.920 --> 00:15:06.700 align:middle line:90% give to things that are bad. 00:15:06.700 --> 00:15:09.630 align:middle line:84% So syntacticians are the opposite of normal people. 00:15:09.630 --> 00:15:12.930 align:middle line:84% When we see things we don't like, we give them a gold star. 00:15:12.930 --> 00:15:13.930 align:middle line:90% Usually, it's not gold. 00:15:13.930 --> 00:15:14.460 align:middle line:90% It's black. 00:15:14.460 --> 00:15:16.085 align:middle line:84% I guess that makes a little more sense. 00:15:16.085 --> 00:15:18.520 align:middle line:90% 00:15:18.520 --> 00:15:21.857 align:middle line:84% This three-way distinction is worth highlighting, 00:15:21.857 --> 00:15:23.940 align:middle line:84% because if I had only given you the first sentence 00:15:23.940 --> 00:15:25.770 align:middle line:84% and the second sentence, "We're going to class," 00:15:25.770 --> 00:15:28.170 align:middle line:84% and "We're class going to," the first sentence is good, 00:15:28.170 --> 00:15:31.500 align:middle line:84% and the second-- and the last sentence is bad, some of you 00:15:31.500 --> 00:15:34.260 align:middle line:84% might be thinking, well, that first sentence, 00:15:34.260 --> 00:15:36.930 align:middle line:84% it's a sentence I've heard lots of times before. 00:15:36.930 --> 00:15:39.600 align:middle line:84% Maybe when I was a small child, I heard my parents say that. 00:15:39.600 --> 00:15:43.460 align:middle line:84% And I heard them say it and I remember it now. 00:15:43.460 --> 00:15:44.490 align:middle line:90% And maybe that's all. 00:15:44.490 --> 00:15:47.190 align:middle line:84% Maybe that's what syntax is, it's the ability to remember 00:15:47.190 --> 00:15:49.800 align:middle line:90% things you've heard people say. 00:15:49.800 --> 00:15:53.370 align:middle line:84% But the existence of the second class of sentences 00:15:53.370 --> 00:15:56.100 align:middle line:90% shows that that's hopeless. 00:15:56.100 --> 00:15:59.610 align:middle line:84% So you're not just remembering things you've heard people say 00:15:59.610 --> 00:16:02.970 align:middle line:84% when you're deciding which things deserve the black star 00:16:02.970 --> 00:16:04.950 align:middle line:90% and which things don't. 00:16:04.950 --> 00:16:07.413 align:middle line:84% You're not just categorizing sentences into sentences 00:16:07.413 --> 00:16:09.330 align:middle line:84% you've heard before and sentences you haven't. 00:16:09.330 --> 00:16:13.140 align:middle line:84% You've got this intuition about which sentences are acceptable. 00:16:13.140 --> 00:16:15.150 align:middle line:84% And it's not just about which sentences 00:16:15.150 --> 00:16:17.010 align:middle line:84% are in your input, which sentences 00:16:17.010 --> 00:16:18.330 align:middle line:90% you've heard people say. 00:16:18.330 --> 00:16:20.220 align:middle line:90% It's something else. 00:16:20.220 --> 00:16:22.230 align:middle line:84% We're going to try to figure out what that is. 00:16:22.230 --> 00:16:24.450 align:middle line:84% But I'm giving you these three sentences 00:16:24.450 --> 00:16:27.960 align:middle line:84% to slay a hypothesis that you might be entertaining. 00:16:27.960 --> 00:16:30.150 align:middle line:84% Stop entertaining that hypothesis. 00:16:30.150 --> 00:16:31.320 align:middle line:90% Make it go home. 00:16:31.320 --> 00:16:33.420 align:middle line:90% It's not a good hypothesis. 00:16:33.420 --> 00:16:34.620 align:middle line:90% It won't do you any good. 00:16:34.620 --> 00:16:35.370 align:middle line:90% Is that clear? 00:16:35.370 --> 00:16:36.787 align:middle line:84% Are people clear on the hypothesis 00:16:36.787 --> 00:16:38.580 align:middle line:90% that I'm attempting to slay? 00:16:38.580 --> 00:16:39.110 align:middle line:90% Yeah? 00:16:39.110 --> 00:16:43.290 align:middle line:84% AUDIENCE: Maybe to [INAUDIBLE] a modified version, what if it's 00:16:43.290 --> 00:16:45.390 align:middle line:84% not particular sentences that we're remembering, 00:16:45.390 --> 00:16:49.833 align:middle line:84% but structures of sentences that we've built up over time? 00:16:49.833 --> 00:16:50.750 align:middle line:90% NORVIN RICHARDS: Yeah. 00:16:50.750 --> 00:16:53.905 align:middle line:84% So if we pursued that-- that's that's much more 00:16:53.905 --> 00:16:55.280 align:middle line:84% sophisticated than the hypothesis 00:16:55.280 --> 00:16:56.450 align:middle line:90% I was trying to slay. 00:16:56.450 --> 00:16:59.783 align:middle line:84% So this is day one of syntax So no, 00:16:59.783 --> 00:17:00.950 align:middle line:90% you're raising a good point. 00:17:00.950 --> 00:17:04.490 align:middle line:84% What if all you're doing is remembering 00:17:04.490 --> 00:17:05.359 align:middle line:90% chunks of sentences? 00:17:05.359 --> 00:17:07.700 align:middle line:84% So maybe you've never heard anybody say, "My anteater 00:17:07.700 --> 00:17:09.170 align:middle line:84% is hula dancing," but maybe you've 00:17:09.170 --> 00:17:11.150 align:middle line:90% heard people say "anteater." 00:17:11.150 --> 00:17:14.839 align:middle line:84% Maybe you've heard people say "My anteater," maybe not. 00:17:14.839 --> 00:17:17.510 align:middle line:84% But at least you've heard people say "My (noun)," 00:17:17.510 --> 00:17:20.000 align:middle line:84% and you've heard people say "anteater," 00:17:20.000 --> 00:17:22.010 align:middle line:90% and "is hula dancing." 00:17:22.010 --> 00:17:22.513 align:middle line:90% Well, yeah. 00:17:22.513 --> 00:17:24.680 align:middle line:84% Maybe you've heard somebody say something like that. 00:17:24.680 --> 00:17:27.230 align:middle line:84% So maybe there are some parts of this that you 00:17:27.230 --> 00:17:28.670 align:middle line:90% could be acquiring that way. 00:17:28.670 --> 00:17:30.605 align:middle line:90% We're going to have to be-- 00:17:30.605 --> 00:17:32.980 align:middle line:84% we're going to have to pursue that hypothesis long enough 00:17:32.980 --> 00:17:36.820 align:middle line:84% to find out exactly what it says, right because what I just 00:17:36.820 --> 00:17:40.030 align:middle line:84% said, which wasn't really the hypothesis, 00:17:40.030 --> 00:17:42.315 align:middle line:84% it was an attempt to represent it, 00:17:42.315 --> 00:17:43.690 align:middle line:84% we're going to have to figure out 00:17:43.690 --> 00:17:47.320 align:middle line:84% how to rule out "We're class going to," because you probably 00:17:47.320 --> 00:17:51.430 align:middle line:84% have heard people say "we're," and "class," 00:17:51.430 --> 00:17:53.680 align:middle line:84% and you've heard people say "going to." 00:17:53.680 --> 00:17:56.170 align:middle line:84% Someone asks, "Where are you going to?" 00:17:56.170 --> 00:17:57.590 align:middle line:90% We've heard people say that. 00:17:57.590 --> 00:18:00.820 align:middle line:84% And so we're going to have to be careful about that 00:18:00.820 --> 00:18:02.710 align:middle line:84% hypothesis to try to figure out how 00:18:02.710 --> 00:18:04.450 align:middle line:90% we could use it to rule and-- 00:18:04.450 --> 00:18:07.240 align:middle line:84% to draw the distinctions that are on the board. 00:18:07.240 --> 00:18:08.260 align:middle line:90% But you're right. 00:18:08.260 --> 00:18:12.730 align:middle line:84% So just to summarize this conversation you and I have 00:18:12.730 --> 00:18:16.400 align:middle line:84% just had, I said, there's a fairly stupid hypothesis, 00:18:16.400 --> 00:18:17.650 align:middle line:90% which says-- 00:18:17.650 --> 00:18:20.020 align:middle line:84% and I'm raising it partly because it's 00:18:20.020 --> 00:18:23.050 align:middle line:84% been seriously entertained in the literature before, which 00:18:23.050 --> 00:18:26.350 align:middle line:84% says, all you're doing is remembering things 00:18:26.350 --> 00:18:28.012 align:middle line:84% people have said, and that's what 00:18:28.012 --> 00:18:29.470 align:middle line:84% distinguishes grammatical sentences 00:18:29.470 --> 00:18:31.030 align:middle line:90% from interpretable sentences. 00:18:31.030 --> 00:18:32.450 align:middle line:90% That's false. 00:18:32.450 --> 00:18:35.080 align:middle line:84% So we have this distinction between sentences 00:18:35.080 --> 00:18:37.360 align:middle line:84% that are grammatical and sentences that 00:18:37.360 --> 00:18:39.112 align:middle line:84% are ungrammatical that isn't just 00:18:39.112 --> 00:18:41.320 align:middle line:84% a list of all the sentences you've ever heard before. 00:18:41.320 --> 00:18:43.362 align:middle line:84% You can take a sentence you've never heard before 00:18:43.362 --> 00:18:45.940 align:middle line:90% and accept it. 00:18:45.940 --> 00:18:47.440 align:middle line:84% You're raising the point there could 00:18:47.440 --> 00:18:49.930 align:middle line:84% be a better version of the stupid hypothesis, one that 00:18:49.930 --> 00:18:51.790 align:middle line:84% said, well, maybe there are some parts 00:18:51.790 --> 00:18:54.610 align:middle line:84% of the sentence that you've heard before, which is true. 00:18:54.610 --> 00:18:56.860 align:middle line:84% We're going to have to be explicit about which 00:18:56.860 --> 00:18:59.918 align:middle line:84% subparts count and what exactly we mean when we say that. 00:18:59.918 --> 00:19:01.960 align:middle line:84% But you're right, there could be a better version 00:19:01.960 --> 00:19:03.830 align:middle line:90% of that hypothesis. 00:19:03.830 --> 00:19:05.020 align:middle line:90% Good point. 00:19:05.020 --> 00:19:06.653 align:middle line:90% Other questions? 00:19:06.653 --> 00:19:08.320 align:middle line:84% Did I successfully answer your question? 00:19:08.320 --> 00:19:10.380 align:middle line:90% Yeah. 00:19:10.380 --> 00:19:11.490 align:middle line:90% OK, all right. 00:19:11.490 --> 00:19:14.760 align:middle line:90% 00:19:14.760 --> 00:19:20.388 align:middle line:84% So let's talk about what's wrong with "We're class going to." 00:19:20.388 --> 00:19:24.820 align:middle line:84% They have several hypotheses about what's wrong with it. 00:19:24.820 --> 00:19:27.890 align:middle line:84% One could be that it doesn't mean anything. 00:19:27.890 --> 00:19:30.040 align:middle line:84% It's a thing people sometimes say. 00:19:30.040 --> 00:19:32.140 align:middle line:84% When I'm trying to explain to people-- 00:19:32.140 --> 00:19:34.810 align:middle line:84% when I'm on an airplane flights and people ask me, 00:19:34.810 --> 00:19:37.790 align:middle line:90% what do you do for a living? 00:19:37.790 --> 00:19:39.490 align:middle line:84% How I answer depends on whether I 00:19:39.490 --> 00:19:42.033 align:middle line:84% feel like talking to the person or not. 00:19:42.033 --> 00:19:43.450 align:middle line:84% So if I feel like talking to them, 00:19:43.450 --> 00:19:46.070 align:middle line:84% then I will tell them that I work on endangered languages, 00:19:46.070 --> 00:19:47.740 align:middle line:84% which is something I do I work with, 00:19:47.740 --> 00:19:49.930 align:middle line:84% languages that are down to their last few speakers, 00:19:49.930 --> 00:19:51.160 align:middle line:90% and try to work with them. 00:19:51.160 --> 00:19:53.110 align:middle line:84% And then sometimes they're interested in that 00:19:53.110 --> 00:19:53.943 align:middle line:90% and they talk to me. 00:19:53.943 --> 00:19:56.410 align:middle line:84% If I would like to get them to leave me 00:19:56.410 --> 00:19:58.278 align:middle line:84% alone so that I can read a book or whatever, 00:19:58.278 --> 00:20:00.070 align:middle line:84% I tell them I'm a theoretical syntactician. 00:20:00.070 --> 00:20:04.310 align:middle line:84% That usually ends the conversation fairly quickly. 00:20:04.310 --> 00:20:07.245 align:middle line:84% But when it doesn't, when they say, "Oh, what's that mean? 00:20:07.245 --> 00:20:08.120 align:middle line:90% What do you work on?" 00:20:08.120 --> 00:20:11.147 align:middle line:84% Then I will say, well, I'm trying to figure out why some 00:20:11.147 --> 00:20:12.980 align:middle line:84% sentences are grammatical and others aren't. 00:20:12.980 --> 00:20:13.938 align:middle line:90% And I'll give examples. 00:20:13.938 --> 00:20:16.450 align:middle line:84% And sometimes they will say, "Oh, but what does that mean?" 00:20:16.450 --> 00:20:17.913 align:middle line:90% "We're class going to." 00:20:17.913 --> 00:20:19.330 align:middle line:84% Maybe that's what's wrong with it. 00:20:19.330 --> 00:20:21.430 align:middle line:90% It doesn't mean anything. 00:20:21.430 --> 00:20:27.220 align:middle line:84% But the problem is we're capable of distinguishing 00:20:27.220 --> 00:20:30.310 align:middle line:84% grammaticality, even in sentences 00:20:30.310 --> 00:20:33.170 align:middle line:90% that don't mean anything. 00:20:33.170 --> 00:20:37.285 align:middle line:84% So this is a famous example of Noam Chomsky's. 00:20:37.285 --> 00:20:39.910 align:middle line:84% In fact, I think if you look him up in like Bartlett's Familiar 00:20:39.910 --> 00:20:42.760 align:middle line:84% Quotations, you'll find this first sentence, 00:20:42.760 --> 00:20:45.585 align:middle line:84% "Colorless green ideas sleep furiously." 00:20:45.585 --> 00:20:46.960 align:middle line:84% He actually offered that sentence 00:20:46.960 --> 00:20:48.670 align:middle line:84% as a part of a pair of sentences. 00:20:48.670 --> 00:20:51.910 align:middle line:84% He wanted people to contrast that sentence 00:20:51.910 --> 00:20:54.610 align:middle line:84% with the same sentence backwards: "Furiously 00:20:54.610 --> 00:20:56.680 align:middle line:90% sleep ideas green colorless." 00:20:56.680 --> 00:20:58.840 align:middle line:90% So consider those two sentences. 00:20:58.840 --> 00:21:03.130 align:middle line:84% And the point is neither of them means anything. 00:21:03.130 --> 00:21:08.420 align:middle line:84% So the first one doesn't mean anything. 00:21:08.420 --> 00:21:11.200 align:middle line:84% And then if you turn your attention to the second one, 00:21:11.200 --> 00:21:13.060 align:middle line:90% it also doesn't mean anything. 00:21:13.060 --> 00:21:15.350 align:middle line:84% But they have a different status. 00:21:15.350 --> 00:21:17.920 align:middle line:84% So the second one doesn't mean anything 00:21:17.920 --> 00:21:21.400 align:middle line:84% and it's ungrammatical, whereas the first one, 00:21:21.400 --> 00:21:24.220 align:middle line:84% it doesn't mean anything, but you feel as though-- 00:21:24.220 --> 00:21:26.830 align:middle line:84% and this is what I'm slowing down right about here 00:21:26.830 --> 00:21:28.720 align:middle line:84% where I say it doesn't mean anything. 00:21:28.720 --> 00:21:30.310 align:middle line:84% Because the reaction people sometimes 00:21:30.310 --> 00:21:31.520 align:middle line:90% have right about here-- 00:21:31.520 --> 00:21:33.680 align:middle line:84% does anyone want-- here, sometimes people say, 00:21:33.680 --> 00:21:34.900 align:middle line:90% "Oh, but look. 00:21:34.900 --> 00:21:40.540 align:middle line:84% Suppose 'colorless' meant 'boring,' and suppose 00:21:40.540 --> 00:21:44.530 align:middle line:84% 'green' meant 'environmentalist,' and suppose 00:21:44.530 --> 00:21:46.570 align:middle line:84% 'sleep' and 'furiously' meant different-- 00:21:46.570 --> 00:21:49.000 align:middle line:84% suppose these words meant something other than what they 00:21:49.000 --> 00:21:51.250 align:middle line:90% mean. 00:21:51.250 --> 00:21:55.318 align:middle line:84% Then the sentence would mean something!"-- which is true. 00:21:55.318 --> 00:21:58.900 align:middle line:84% But it's another way of saying the same point. 00:21:58.900 --> 00:22:02.710 align:middle line:84% Yeah, the first sentence, "Colorless green ideas 00:22:02.710 --> 00:22:05.650 align:middle line:84% sleep furiously," is meaningless if you don't mess 00:22:05.650 --> 00:22:07.630 align:middle line:84% with the meanings of the sentences. 00:22:07.630 --> 00:22:11.600 align:middle line:84% But it obeys the rules for how words can be combined. 00:22:11.600 --> 00:22:13.180 align:middle line:84% If the words meant something else, 00:22:13.180 --> 00:22:15.800 align:middle line:90% the sentence would be fine. 00:22:15.800 --> 00:22:17.320 align:middle line:84% We can have English sentences that 00:22:17.320 --> 00:22:21.380 align:middle line:84% consist of two adjectives modifying a noun. 00:22:21.380 --> 00:22:24.130 align:middle line:84% And then there's a verb, and then there's an adverb. 00:22:24.130 --> 00:22:32.560 align:middle line:84% Not that one, but "Big green monsters snore loudly," 00:22:32.560 --> 00:22:33.370 align:middle line:90% that would be fine. 00:22:33.370 --> 00:22:36.110 align:middle line:90% 00:22:36.110 --> 00:22:40.480 align:middle line:84% So when the person on the airplane flight next to me 00:22:40.480 --> 00:22:44.500 align:middle line:84% says, "Oh, what's wrong with 'We're class going to' 00:22:44.500 --> 00:22:47.133 align:middle line:84% is that it's a meaningless sentence," if I'm 00:22:47.133 --> 00:22:49.550 align:middle line:84% really, really desperately trying to end the conversation, 00:22:49.550 --> 00:22:51.370 align:middle line:84% I bring out these kinds of pairs. 00:22:51.370 --> 00:22:56.140 align:middle line:90% So it's not about meaning. 00:22:56.140 --> 00:23:00.130 align:middle line:84% We have this intuition that there are sentences that are OK 00:23:00.130 --> 00:23:03.070 align:middle line:84% and sentences that are bad, which is separable from 00:23:03.070 --> 00:23:06.657 align:middle line:84% our intuition about what means something and what doesn't. 00:23:06.657 --> 00:23:08.740 align:middle line:84% I've just been asserting things about our feelings 00:23:08.740 --> 00:23:09.460 align:middle line:90% about these sentences. 00:23:09.460 --> 00:23:11.500 align:middle line:84% Do people have this feeling about these sentences? 00:23:11.500 --> 00:23:13.333 align:middle line:84% First, that they're meaningless, and second, 00:23:13.333 --> 00:23:17.530 align:middle line:84% that the first one is OK and the second one is bad? 00:23:17.530 --> 00:23:18.395 align:middle line:90% Yeah, Raquel? 00:23:18.395 --> 00:23:20.270 align:middle line:84% AUDIENCE: I can't remember the word for this, 00:23:20.270 --> 00:23:22.240 align:middle line:84% but you were saying that certain types of words 00:23:22.240 --> 00:23:24.640 align:middle line:84% are categories that you can add more words to, 00:23:24.640 --> 00:23:27.525 align:middle line:84% even if you don't know what they like modifiers 00:23:27.525 --> 00:23:28.400 align:middle line:90% and things like that. 00:23:28.400 --> 00:23:31.987 align:middle line:84% And like, "This house is very [blope?]" That's grammatical, 00:23:31.987 --> 00:23:34.570 align:middle line:84% even if you don't know what it means, or you can't make that-- 00:23:34.570 --> 00:23:35.612 align:middle line:90% NORVIN RICHARDS: Oh, yes. 00:23:35.612 --> 00:23:36.640 align:middle line:90% Yes. 00:23:36.640 --> 00:23:41.980 align:middle line:84% We were talking about open class and closed class morphemes. 00:23:41.980 --> 00:23:43.630 align:middle line:84% So Jabberwocky is a poem that you 00:23:43.630 --> 00:23:47.350 align:middle line:84% can write, changing all the lexical items 00:23:47.350 --> 00:23:49.940 align:middle line:84% to nonsense words, but you couldn't 00:23:49.940 --> 00:23:51.190 align:middle line:90% do that with functional items. 00:23:51.190 --> 00:23:53.110 align:middle line:90% Yeah, that's right. 00:23:53.110 --> 00:23:57.100 align:middle line:84% AUDIENCE: So maybe in situations where something doesn't really 00:23:57.100 --> 00:24:00.550 align:middle line:84% make any sense, your brain still knows that OK, 00:24:00.550 --> 00:24:02.980 align:middle line:84% maybe it almost like that kind of class 00:24:02.980 --> 00:24:04.480 align:middle line:84% where you're like, well, I mean, you 00:24:04.480 --> 00:24:06.340 align:middle line:84% can fit like an adjective like this-- 00:24:06.340 --> 00:24:06.660 align:middle line:90% NORVIN RICHARDS: Yep. 00:24:06.660 --> 00:24:08.702 align:middle line:84% AUDIENCE: --even if it means something ridiculous 00:24:08.702 --> 00:24:12.273 align:middle line:84% right here, and it sounds OK even if it's [INAUDIBLE].. 00:24:12.273 --> 00:24:13.190 align:middle line:90% NORVIN RICHARDS: Yeah. 00:24:13.190 --> 00:24:14.545 align:middle line:90% That's a nice way to put it. 00:24:14.545 --> 00:24:15.920 align:middle line:84% I guess this is similar to what I 00:24:15.920 --> 00:24:18.650 align:middle line:84% was trying to say about a reaction I sometimes 00:24:18.650 --> 00:24:20.995 align:middle line:84% get to this sentence, which is, you 00:24:20.995 --> 00:24:22.940 align:middle line:84% say "Colorless green ideas sleep furiously" 00:24:22.940 --> 00:24:24.360 align:middle line:90% doesn't mean anything. 00:24:24.360 --> 00:24:25.920 align:middle line:84% And people will sometimes say, "Well, 00:24:25.920 --> 00:24:27.650 align:middle line:84% but if these words meant something else, 00:24:27.650 --> 00:24:30.590 align:middle line:84% then it would be OK," which is true. 00:24:30.590 --> 00:24:31.130 align:middle line:90% Yeah? 00:24:31.130 --> 00:24:32.505 align:middle line:84% AUDIENCE: I wonder if it just has 00:24:32.505 --> 00:24:44.413 align:middle line:90% to do with the idea [INAUDIBLE] 00:24:44.413 --> 00:24:45.330 align:middle line:90% NORVIN RICHARDS: Yeah. 00:24:45.330 --> 00:24:47.560 align:middle line:84% I think I may not have heard all of that, 00:24:47.560 --> 00:24:50.070 align:middle line:84% but the idea was the first sentence, 00:24:50.070 --> 00:24:52.980 align:middle line:84% it's clear what "colorless" and "green" are trying to do. 00:24:52.980 --> 00:24:54.990 align:middle line:84% They're both trying to modify ideas. 00:24:54.990 --> 00:24:56.640 align:middle line:90% Is that right? 00:24:56.640 --> 00:24:58.930 align:middle line:84% And it's clear what "furiously" is trying to do. 00:24:58.930 --> 00:25:01.290 align:middle line:84% It's an adverb and it's trying to modify "sleep." 00:25:01.290 --> 00:25:02.700 align:middle line:84% When I say it's meaningless, what 00:25:02.700 --> 00:25:05.550 align:middle line:84% I mean is you can't sleep furiously, right? 00:25:05.550 --> 00:25:08.130 align:middle line:84% And things can't be both colorless and green. 00:25:08.130 --> 00:25:11.520 align:middle line:84% And if they could, well, ideas couldn't have those things. 00:25:11.520 --> 00:25:12.630 align:middle line:90% They're abstract, right? 00:25:12.630 --> 00:25:17.520 align:middle line:84% That's the sense in which this is a meaningless sentence. 00:25:17.520 --> 00:25:21.390 align:middle line:84% I think you're raising the point, which is a good point, 00:25:21.390 --> 00:25:24.600 align:middle line:84% yeah, it's meaningless, but the words in it 00:25:24.600 --> 00:25:26.730 align:middle line:84% are fitting together the way they should. 00:25:26.730 --> 00:25:29.130 align:middle line:84% The adjectives are going where adjectives go. 00:25:29.130 --> 00:25:32.160 align:middle line:84% They go before the noun, that's what they're supposed to do. 00:25:32.160 --> 00:25:36.270 align:middle line:84% And in the second sentence, they're not doing that. 00:25:36.270 --> 00:25:38.890 align:middle line:84% Is that the relation to the point that you're making? 00:25:38.890 --> 00:25:39.900 align:middle line:90% Yeah. 00:25:39.900 --> 00:25:40.650 align:middle line:90% Yeah? 00:25:40.650 --> 00:25:44.376 align:middle line:84% So that's the intuition that we have, 00:25:44.376 --> 00:25:48.450 align:middle line:84% that it's possible to have feelings about sentences 00:25:48.450 --> 00:25:51.790 align:middle line:84% of the form, well, I don't know what this is supposed to mean, 00:25:51.790 --> 00:25:54.270 align:middle line:84% but the parts are all in the right place. 00:25:54.270 --> 00:25:57.330 align:middle line:84% We are in the syntax part of this class. 00:25:57.330 --> 00:26:00.180 align:middle line:84% It's all about parts being in the right places. 00:26:00.180 --> 00:26:01.990 align:middle line:84% We will eventually do the semantics part, 00:26:01.990 --> 00:26:03.415 align:middle line:90% which is about meaning. 00:26:03.415 --> 00:26:05.790 align:middle line:84% But the point is that it's possible to study these things 00:26:05.790 --> 00:26:07.660 align:middle line:90% independently of each other. 00:26:07.660 --> 00:26:10.680 align:middle line:90% So completely independently. 00:26:10.680 --> 00:26:14.340 align:middle line:84% So when I was showing you "unlockable," 00:26:14.340 --> 00:26:16.590 align:middle line:84% what we were really doing was morphology, 00:26:16.590 --> 00:26:18.240 align:middle line:84% but we were also talking about meaning. 00:26:18.240 --> 00:26:20.640 align:middle line:84% We were interested in the fact that word 00:26:20.640 --> 00:26:21.820 align:middle line:90% had two different meanings. 00:26:21.820 --> 00:26:25.000 align:middle line:84% We talked about the fact that it meant different things. 00:26:25.000 --> 00:26:26.880 align:middle line:84% So we get to use semantics as kind of a probe 00:26:26.880 --> 00:26:28.600 align:middle line:84% into what's happening in morphology. 00:26:28.600 --> 00:26:30.930 align:middle line:84% We'll do things like that in syntax. 00:26:30.930 --> 00:26:33.270 align:middle line:84% But our feelings about whether sentences 00:26:33.270 --> 00:26:35.790 align:middle line:84% are grammatical or not, or acceptable or not, 00:26:35.790 --> 00:26:38.170 align:middle line:84% are separable from our feelings about what they mean. 00:26:38.170 --> 00:26:39.420 align:middle line:90% That's the point, if anything. 00:26:39.420 --> 00:26:45.920 align:middle line:90% 00:26:45.920 --> 00:26:50.087 align:middle line:84% The reverse, there are sentences that are ungrammatical, but-- 00:26:50.087 --> 00:26:51.670 align:middle line:84% that are meaningless, but grammatical, 00:26:51.670 --> 00:26:54.040 align:middle line:84% like "Colorless green ideas sleep furiously." 00:26:54.040 --> 00:26:56.950 align:middle line:84% On the flip side, there are sentences 00:26:56.950 --> 00:26:59.213 align:middle line:84% that are meaningful, but ungrammatical, 00:26:59.213 --> 00:27:01.630 align:middle line:84% the sentences where it's very clear what they should mean, 00:27:01.630 --> 00:27:04.010 align:middle line:84% but you just can't say it that way. 00:27:04.010 --> 00:27:08.450 align:middle line:84% Here's a quadruple of examples that's meant to show you that. 00:27:08.450 --> 00:27:11.110 align:middle line:84% So you can say, "Put the sweater on," you can say, 00:27:11.110 --> 00:27:12.940 align:middle line:84% "I put on the sweater," you can say, 00:27:12.940 --> 00:27:17.050 align:middle line:84% "I put it on," you cannot say, "I put on it." 00:27:17.050 --> 00:27:18.820 align:middle line:84% The meaning is not the problem here. 00:27:18.820 --> 00:27:22.060 align:middle line:84% It's clear what "I put on it" would mean. 00:27:22.060 --> 00:27:24.970 align:middle line:84% But there are facts about how English pronouns work 00:27:24.970 --> 00:27:27.550 align:middle line:84% and how English particles work that mean 00:27:27.550 --> 00:27:29.530 align:middle line:90% that you don't get to say that. 00:27:29.530 --> 00:27:31.280 align:middle line:90% It's not a meaning thing. 00:27:31.280 --> 00:27:32.890 align:middle line:90% It's something about syntax. 00:27:32.890 --> 00:27:35.830 align:middle line:84% It's something about how these parts get to combine. 00:27:35.830 --> 00:27:39.220 align:middle line:84% You want to try to understand that. 00:27:39.220 --> 00:27:40.930 align:middle line:84% The only point of these few slides 00:27:40.930 --> 00:27:44.080 align:middle line:84% has been it's possible to study syntax independently 00:27:44.080 --> 00:27:46.900 align:middle line:84% of meaning, where by independently, I just 00:27:46.900 --> 00:27:49.150 align:middle line:84% mean the facts of syntax don't just 00:27:49.150 --> 00:27:50.770 align:middle line:90% reduce to facts about meaning. 00:27:50.770 --> 00:27:52.420 align:middle line:84% They also don't just reduce to facts 00:27:52.420 --> 00:27:54.372 align:middle line:90% about what you've heard before. 00:27:54.372 --> 00:27:56.080 align:middle line:84% That's what I've been trying to show you. 00:27:56.080 --> 00:27:59.790 align:middle line:90% 00:27:59.790 --> 00:28:02.180 align:middle line:84% Here's another thing you might think about what's wrong 00:28:02.180 --> 00:28:04.190 align:middle line:90% with "We're class going to." 00:28:04.190 --> 00:28:06.380 align:middle line:90% It ends in a preposition. 00:28:06.380 --> 00:28:09.020 align:middle line:84% Where any of you taught in school you must not end 00:28:09.020 --> 00:28:11.500 align:middle line:90% sentences with prepositions? 00:28:11.500 --> 00:28:12.400 align:middle line:90% Some of you were-- 00:28:12.400 --> 00:28:13.380 align:middle line:90% AUDIENCE: [INAUDIBLE] 00:28:13.380 --> 00:28:14.070 align:middle line:90% NORVIN RICHARDS: Yeah. 00:28:14.070 --> 00:28:14.880 align:middle line:90% Some of you were not. 00:28:14.880 --> 00:28:15.640 align:middle line:90% That's kind of interesting. 00:28:15.640 --> 00:28:16.308 align:middle line:90% I was. 00:28:16.308 --> 00:28:18.100 align:middle line:84% I was beaten by English teachers for ending 00:28:18.100 --> 00:28:18.960 align:middle line:90% a sentence-- that's not true. 00:28:18.960 --> 00:28:19.690 align:middle line:90% I was not beaten. 00:28:19.690 --> 00:28:22.050 align:middle line:84% I was spoken to harshly by English teachers 00:28:22.050 --> 00:28:24.270 align:middle line:84% for ending sentences with prepositions. 00:28:24.270 --> 00:28:26.730 align:middle line:84% Do you know why we were told not to end sentences 00:28:26.730 --> 00:28:28.518 align:middle line:90% with prepositions? 00:28:28.518 --> 00:28:30.740 align:middle line:90% AUDIENCE: It's reductive? 00:28:30.740 --> 00:28:37.460 align:middle line:84% Because I guess-- because where are we going to 00:28:37.460 --> 00:28:40.955 align:middle line:90% could just be a [INAUDIBLE]. 00:28:40.955 --> 00:28:42.830 align:middle line:84% NORVIN RICHARDS: So there are some cases like 00:28:42.830 --> 00:28:47.180 align:middle line:84% that but look there are also cases like, "Who are you 00:28:47.180 --> 00:28:52.760 align:middle line:90% talking to" where-- 00:28:52.760 --> 00:28:55.610 align:middle line:84% sorry, where-- who am I apologizing to? 00:28:55.610 --> 00:28:58.280 align:middle line:90% The chalk, I guess. 00:28:58.280 --> 00:29:02.450 align:middle line:84% "Who are you talking to" where, without the "to," 00:29:02.450 --> 00:29:04.850 align:middle line:84% it doesn't have to be about redundancy, I don't think. 00:29:04.850 --> 00:29:07.110 align:middle line:84% Is this a sentence that you've ever heard anyone say? 00:29:07.110 --> 00:29:07.910 align:middle line:90% Is this OK? 00:29:07.910 --> 00:29:10.940 align:middle line:84% Yeah, this is, I think, pretty good English. 00:29:10.940 --> 00:29:13.220 align:middle line:84% So when our English teachers were telling us 00:29:13.220 --> 00:29:16.010 align:middle line:84% not to end sentences with prepositions, 00:29:16.010 --> 00:29:18.695 align:middle line:84% we should have asked them, what are you talking about? 00:29:18.695 --> 00:29:22.200 align:middle line:90% 00:29:22.200 --> 00:29:25.490 align:middle line:84% Because the fact is that English speakers 00:29:25.490 --> 00:29:28.400 align:middle line:84% end sentences with prepositions every day. 00:29:28.400 --> 00:29:32.150 align:middle line:84% Do you know why your teacher told you not to do that? 00:29:32.150 --> 00:29:32.930 align:middle line:90% Yes? 00:29:32.930 --> 00:29:35.680 align:middle line:84% AUDIENCE: Maybe it's because the object isn't clear? 00:29:35.680 --> 00:29:37.810 align:middle line:84% NORVIN RICHARDS: Well-- but is it? 00:29:37.810 --> 00:29:39.550 align:middle line:90% I mean, it's kind of-- 00:29:39.550 --> 00:29:41.900 align:middle line:84% if I ask you, who are you talking to? 00:29:41.900 --> 00:29:43.900 align:middle line:84% There's a sense in which the object isn't clear. 00:29:43.900 --> 00:29:46.360 align:middle line:84% I'm asking you who the object is, right? 00:29:46.360 --> 00:29:50.558 align:middle line:84% So it shouldn't be any harder than, who are you-- 00:29:50.558 --> 00:29:51.850 align:middle line:90% AUDIENCE: Who are you talking-- 00:29:51.850 --> 00:29:54.500 align:middle line:84% NORVIN RICHARDS: --describing, yeah. 00:29:54.500 --> 00:29:55.000 align:middle line:90% Yeah? 00:29:55.000 --> 00:29:58.180 align:middle line:84% AUDIENCE: Is it because the "who" is supposed 00:29:58.180 --> 00:30:02.450 align:middle line:84% to be the direct object of "you are talking to" this person, 00:30:02.450 --> 00:30:04.460 align:middle line:84% so it's like, "to whom are you talking"? 00:30:04.460 --> 00:30:06.138 align:middle line:90% NORVIN RICHARDS: So yeah. 00:30:06.138 --> 00:30:07.930 align:middle line:84% What your English teacher wanted you to say 00:30:07.930 --> 00:30:09.222 align:middle line:90% was, "To whom are you talking?" 00:30:09.222 --> 00:30:15.700 align:middle line:90% 00:30:15.700 --> 00:30:21.600 align:middle line:84% which many English speakers are capable of saying, possibly 00:30:21.600 --> 00:30:25.470 align:middle line:84% because their English teachers frightened them into it. 00:30:25.470 --> 00:30:27.483 align:middle line:84% But it's not my go-to way to say this. 00:30:27.483 --> 00:30:28.650 align:middle line:90% I don't know about you guys. 00:30:28.650 --> 00:30:32.280 align:middle line:90% 00:30:32.280 --> 00:30:34.950 align:middle line:84% So here, "who," the question word, 00:30:34.950 --> 00:30:36.750 align:middle line:84% is up here at the beginning by itself. 00:30:36.750 --> 00:30:40.560 align:middle line:84% Here, "to whom" is part of this phrase that's at the beginning. 00:30:40.560 --> 00:30:46.020 align:middle line:84% But question, why does "to" have to come along, according 00:30:46.020 --> 00:30:47.400 align:middle line:90% to your English teacher? 00:30:47.400 --> 00:30:47.910 align:middle line:90% Yes? 00:30:47.910 --> 00:30:48.930 align:middle line:84% AUDIENCE: Is this another example 00:30:48.930 --> 00:30:50.280 align:middle line:84% because that's how it's done in Latin? 00:30:50.280 --> 00:30:50.535 align:middle line:90% NORVIN RICHARDS: Yes. 00:30:50.535 --> 00:30:51.415 align:middle line:90% AUDIENCE: [INAUDIBLE] 00:30:51.415 --> 00:30:52.290 align:middle line:90% NORVIN RICHARDS: Yes. 00:30:52.290 --> 00:30:52.962 align:middle line:90% Yes. 00:30:52.962 --> 00:30:54.420 align:middle line:84% Your English teacher told you to do 00:30:54.420 --> 00:30:59.160 align:middle line:84% that because Latin, actually, among many other languages, 00:30:59.160 --> 00:31:01.390 align:middle line:90% doesn't allow you to do this. 00:31:01.390 --> 00:31:03.750 align:middle line:90% You have to do this. 00:31:03.750 --> 00:31:06.780 align:middle line:84% English is quite rare in being able to do this. 00:31:06.780 --> 00:31:09.870 align:middle line:84% Most of the languages of the world can't. 00:31:09.870 --> 00:31:14.100 align:middle line:84% And some time in the 15th, 16th century, 00:31:14.100 --> 00:31:17.280 align:middle line:84% a number of grammarians decided that English 00:31:17.280 --> 00:31:19.830 align:middle line:84% would be way cooler if it were more like Latin, 00:31:19.830 --> 00:31:23.400 align:middle line:84% and so they began declaring that it was. 00:31:23.400 --> 00:31:26.010 align:middle line:84% And that's why your English teacher told you 00:31:26.010 --> 00:31:30.300 align:middle line:84% that you can't say this, which is ridiculous. 00:31:30.300 --> 00:31:32.610 align:middle line:84% We should be proud of the fact that we can say this, 00:31:32.610 --> 00:31:33.960 align:middle line:90% because, as I say, it's rare. 00:31:33.960 --> 00:31:35.250 align:middle line:90% Most languages can't do this. 00:31:35.250 --> 00:31:38.130 align:middle line:84% We should have this on our flag or something like that. 00:31:38.130 --> 00:31:41.430 align:middle line:84% There should be a stranded preposition, a preposition 00:31:41.430 --> 00:31:43.620 align:middle line:90% at the end of the sentence. 00:31:43.620 --> 00:31:48.540 align:middle line:84% So yeah, no reason English has to be like Latin. 00:31:48.540 --> 00:31:51.210 align:middle line:84% Or like I say, a zillion other languages. 00:31:51.210 --> 00:31:54.570 align:middle line:84% Most of the languages of Europe, French, or German, 00:31:54.570 --> 00:31:57.240 align:middle line:84% or Italian, or whatever, you can't leave prepositions 00:31:57.240 --> 00:31:58.927 align:middle line:90% at the ends of sentences. 00:31:58.927 --> 00:32:01.260 align:middle line:84% But those teachers don't have to tell their children not 00:32:01.260 --> 00:32:03.540 align:middle line:84% to say things like, who are you talking to, 00:32:03.540 --> 00:32:06.270 align:middle line:84% because they literally can't, and none of those kids 00:32:06.270 --> 00:32:07.600 align:middle line:90% would ever do that. 00:32:07.600 --> 00:32:08.835 align:middle line:90% But in English, we can. 00:32:08.835 --> 00:32:09.960 align:middle line:90% We should be proud of that. 00:32:09.960 --> 00:32:13.080 align:middle line:90% 00:32:13.080 --> 00:32:16.710 align:middle line:84% So that's not the problem with "We're class going to." 00:32:16.710 --> 00:32:19.650 align:middle line:84% It does end in a preposition, but many perfectly fine 00:32:19.650 --> 00:32:22.920 align:middle line:84% English sentences end in prepositions. 00:32:22.920 --> 00:32:24.930 align:middle line:84% There's a distinction-- yeah, there's 00:32:24.930 --> 00:32:26.730 align:middle line:84% the example "Who are you talking to?" 00:32:26.730 --> 00:32:28.147 align:middle line:84% There's a distinction that's worth 00:32:28.147 --> 00:32:32.490 align:middle line:84% drawing between what's called prescriptive and descriptive 00:32:32.490 --> 00:32:33.910 align:middle line:90% studies of grammar. 00:32:33.910 --> 00:32:36.450 align:middle line:84% So what we are doing in this class 00:32:36.450 --> 00:32:39.270 align:middle line:84% is trying to figure out what people actually say, 00:32:39.270 --> 00:32:43.470 align:middle line:84% what the rules are for putting sentences together in English. 00:32:43.470 --> 00:32:47.160 align:middle line:84% There are other kinds of things that people say about how 00:32:47.160 --> 00:32:49.240 align:middle line:90% English should be spoken. 00:32:49.240 --> 00:32:52.140 align:middle line:84% We're not going to talk about that stuff, except to mock it 00:32:52.140 --> 00:32:53.880 align:middle line:90% the way I did just now. 00:32:53.880 --> 00:32:57.703 align:middle line:84% So prescriptive grammar is the study of rules 00:32:57.703 --> 00:32:59.370 align:middle line:84% that your teachers might have taught you 00:32:59.370 --> 00:33:01.860 align:middle line:84% in school about how to speak, some of which, 00:33:01.860 --> 00:33:04.200 align:middle line:84% just to stop mocking it for a second, 00:33:04.200 --> 00:33:06.660 align:middle line:84% your teachers might have tried to tell you 00:33:06.660 --> 00:33:09.240 align:middle line:84% things that would genuinely improve 00:33:09.240 --> 00:33:12.480 align:middle line:84% the quality of your writing, like get rid of ambiguities 00:33:12.480 --> 00:33:13.290 align:middle line:90% of various kinds. 00:33:13.290 --> 00:33:17.100 align:middle line:84% So if you're now mentally composing nasty messages 00:33:17.100 --> 00:33:18.970 align:middle line:84% to your English teachers, don't send them. 00:33:18.970 --> 00:33:20.633 align:middle line:84% They had your best interests at heart, 00:33:20.633 --> 00:33:22.050 align:middle line:84% and they probably taught you a lot 00:33:22.050 --> 00:33:23.430 align:middle line:90% of things that were valuable. 00:33:23.430 --> 00:33:25.140 align:middle line:84% But they also taught you some things 00:33:25.140 --> 00:33:29.195 align:middle line:84% that became popular around the 15th, 16th century 00:33:29.195 --> 00:33:31.320 align:middle line:84% because people thought that English would be better 00:33:31.320 --> 00:33:32.760 align:middle line:90% if it were more like Latin. 00:33:32.760 --> 00:33:36.300 align:middle line:84% And so we're not going to try to improve your writing 00:33:36.300 --> 00:33:38.790 align:middle line:84% in this class, except insofar as the writing 00:33:38.790 --> 00:33:41.160 align:middle line:90% advisors can do that. 00:33:41.160 --> 00:33:44.340 align:middle line:84% We're not going to be talking about prescriptive studies 00:33:44.340 --> 00:33:45.570 align:middle line:90% of English grammar. 00:33:45.570 --> 00:33:47.250 align:middle line:84% We're not going to talk about how you should speak, 00:33:47.250 --> 00:33:48.210 align:middle line:90% or how you should write. 00:33:48.210 --> 00:33:49.890 align:middle line:84% We're just going to talk about how you do. 00:33:49.890 --> 00:33:52.432 align:middle line:84% So this is going to be a study of descriptive grammar and not 00:33:52.432 --> 00:33:53.760 align:middle line:90% prescriptive. 00:33:53.760 --> 00:33:55.318 align:middle line:90% Yeah? 00:33:55.318 --> 00:33:56.860 align:middle line:84% AUDIENCE: So for that second sentence 00:33:56.860 --> 00:33:59.110 align:middle line:84% on the board, "What are you talking about?" 00:33:59.110 --> 00:34:00.360 align:middle line:90% NORVIN RICHARDS: Oh, this one? 00:34:00.360 --> 00:34:00.886 align:middle line:90% Yeah? 00:34:00.886 --> 00:34:05.105 align:middle line:84% AUDIENCE: How would you answer that such that something 00:34:05.105 --> 00:34:08.030 align:middle line:84% [? throws ?] the word [? around. ?] Because "About 00:34:08.030 --> 00:34:09.806 align:middle line:90% what are you talking?" 00:34:09.806 --> 00:34:10.659 align:middle line:90% is-- 00:34:10.659 --> 00:34:13.326 align:middle line:84% NORVIN RICHARDS: I can't do that in English, "About what are you 00:34:13.326 --> 00:34:15.432 align:middle line:90% talking?" 00:34:15.432 --> 00:34:18.550 align:middle line:84% AUDIENCE: That sounds wrong, actually. 00:34:18.550 --> 00:34:21.480 align:middle line:84% Whereas "To whom are you talking?" 00:34:21.480 --> 00:34:23.572 align:middle line:84% sounds fancy and snobbish, but still right. 00:34:23.572 --> 00:34:24.989 align:middle line:84% NORVIN RICHARDS: Yeah, yeah, yeah. 00:34:24.989 --> 00:34:27.389 align:middle line:84% No, you're raising an interesting distinction. 00:34:27.389 --> 00:34:30.840 align:middle line:84% So in many languages, most languages, including Latin, you 00:34:30.840 --> 00:34:32.610 align:middle line:84% have to say, "About what are you talking?" 00:34:32.610 --> 00:34:34.568 align:middle line:84% and you have to say, "To whom are you talking?" 00:34:34.568 --> 00:34:37.215 align:middle line:84% So these are languages in which you never leave a preposition 00:34:37.215 --> 00:34:38.340 align:middle line:90% at the end of the sentence. 00:34:38.340 --> 00:34:40.594 align:middle line:84% You always bring it along with the question word 00:34:40.594 --> 00:34:41.969 align:middle line:84% to the beginning of the sentence. 00:34:41.969 --> 00:34:43.886 align:middle line:84% But you're raising a really interesting point. 00:34:43.886 --> 00:34:45.719 align:middle line:84% In English, there's this distinction 00:34:45.719 --> 00:34:50.280 align:middle line:84% between the examples where leaving a preposition behind 00:34:50.280 --> 00:34:52.840 align:middle line:84% is what you prefer, "Who are you talking to" 00:34:52.840 --> 00:34:55.110 align:middle line:84% But you can kind of say, "To whom are you talking?" 00:34:55.110 --> 00:34:58.380 align:middle line:84% And others, were "About what are you talking?" 00:34:58.380 --> 00:35:01.050 align:middle line:84% Do other people have this intuition, that "About what are 00:35:01.050 --> 00:35:04.080 align:middle line:84% you talking?" is worse than "To whom are you talking?" 00:35:04.080 --> 00:35:07.560 align:middle line:84% I have that feeling, too, I think. 00:35:07.560 --> 00:35:10.300 align:middle line:90% There are other examples. 00:35:10.300 --> 00:35:12.885 align:middle line:84% There are examples that are really quite bad. 00:35:12.885 --> 00:35:18.630 align:middle line:90% 00:35:18.630 --> 00:35:20.130 align:middle line:84% And other examples which get better. 00:35:20.130 --> 00:35:32.445 align:middle line:84% So things like "We left despite her warnings." 00:35:32.445 --> 00:35:36.540 align:middle line:90% 00:35:36.540 --> 00:35:38.730 align:middle line:84% And then consider two kinds of questions 00:35:38.730 --> 00:35:39.940 align:middle line:90% you could ask about that. 00:35:39.940 --> 00:35:41.325 align:middle line:90% "What did you leave despite?" 00:35:41.325 --> 00:35:49.570 align:middle line:90% 00:35:49.570 --> 00:35:51.730 align:middle line:84% And "Despite what did you leave?" 00:35:51.730 --> 00:35:59.890 align:middle line:90% 00:35:59.890 --> 00:36:02.490 align:middle line:84% Is either of those acceptable at all? 00:36:02.490 --> 00:36:03.810 align:middle line:90% "What did you leave despite?" 00:36:03.810 --> 00:36:05.280 align:middle line:84% and "Despite what did you leave?" 00:36:05.280 --> 00:36:08.030 align:middle line:84% Who prefers "What did you leave despite?" 00:36:08.030 --> 00:36:10.220 align:middle line:84% Who prefers "Despite what did you leave?" 00:36:10.220 --> 00:36:13.610 align:middle line:84% Who would do anything to avoid saying either of these things? 00:36:13.610 --> 00:36:14.670 align:middle line:90% Yes. 00:36:14.670 --> 00:36:15.380 align:middle line:90% Yeah, yeah, yeah. 00:36:15.380 --> 00:36:20.730 align:middle line:84% So here's another example of one sentence that's quite bad 00:36:20.730 --> 00:36:24.040 align:middle line:84% and another sentence that's only bad-ish. 00:36:24.040 --> 00:36:26.690 align:middle line:84% You might want to try to understand what's going on. 00:36:26.690 --> 00:36:28.850 align:middle line:84% So there's a fruitful area of research here. 00:36:28.850 --> 00:36:30.380 align:middle line:90% OK, English doesn't-- 00:36:30.380 --> 00:36:32.060 align:middle line:84% English is happy to leave prepositions 00:36:32.060 --> 00:36:33.200 align:middle line:90% at the ends of sentences. 00:36:33.200 --> 00:36:36.000 align:middle line:84% But in which cases is it happy to do the Latin thing? 00:36:36.000 --> 00:36:37.870 align:middle line:84% And in some cases, it's happier than others. 00:36:37.870 --> 00:36:39.245 align:middle line:84% It's interesting to try to study. 00:36:39.245 --> 00:36:47.650 align:middle line:90% 00:36:47.650 --> 00:36:49.190 align:middle line:90% Another distinction to make. 00:36:49.190 --> 00:36:50.950 align:middle line:84% So we've drawn this distinction now 00:36:50.950 --> 00:36:54.640 align:middle line:84% between meaningless on the one hand 00:36:54.640 --> 00:36:56.320 align:middle line:90% and ungrammatical on the other. 00:36:56.320 --> 00:36:58.750 align:middle line:84% That sentence can be both meaningless and ungrammatical, 00:36:58.750 --> 00:37:00.708 align:middle line:84% but it can also be meaningless and grammatical, 00:37:00.708 --> 00:37:04.202 align:middle line:84% like "Colorless green ideas sleep furiously." 00:37:04.202 --> 00:37:05.410 align:middle line:90% We've drawn that distinction. 00:37:05.410 --> 00:37:09.280 align:middle line:84% We've drawn a distinction between prescriptive and 00:37:09.280 --> 00:37:10.378 align:middle line:90% descriptive statements. 00:37:10.378 --> 00:37:11.920 align:middle line:84% So prescriptive statements are things 00:37:11.920 --> 00:37:15.220 align:middle line:84% like "Don't end sentences with prepositions," which are-- 00:37:15.220 --> 00:37:17.290 align:middle line:90% how shall I say this?-- false. 00:37:17.290 --> 00:37:19.870 align:middle line:84% Not a good description of what English speakers actually do. 00:37:19.870 --> 00:37:22.085 align:middle line:90% They are aspirations. 00:37:22.085 --> 00:37:23.210 align:middle line:90% We're not going to do that. 00:37:23.210 --> 00:37:27.460 align:middle line:84% We're just going to study what English speakers do. 00:37:27.460 --> 00:37:29.080 align:middle line:84% So here's another useful distinction. 00:37:29.080 --> 00:37:33.430 align:middle line:84% It's sometimes called competence versus performance. 00:37:33.430 --> 00:37:37.130 align:middle line:84% Imagine that I am standing up here talking to you. 00:37:37.130 --> 00:37:40.220 align:middle line:84% So far, this should be easy to imagine. 00:37:40.220 --> 00:37:41.410 align:middle line:90% Except I'm not standing. 00:37:41.410 --> 00:37:45.370 align:middle line:84% Imagine that I'm standing up here talking to you and I say, 00:37:45.370 --> 00:37:49.107 align:middle line:84% "This is the--" and then I Inhale a fly. 00:37:49.107 --> 00:37:51.190 align:middle line:84% So imagine that-- oh, I don't have to wear a mask. 00:37:51.190 --> 00:37:53.710 align:middle line:90% I keep forgetting that. 00:37:53.710 --> 00:37:56.380 align:middle line:84% Imagine that I take off my mask and I Inhale a fly. 00:37:56.380 --> 00:37:58.780 align:middle line:84% So I say, "This is the--" and then I stop. 00:37:58.780 --> 00:38:01.990 align:middle line:84% I'm like [IMITATES COUGHING] and I stop right there. 00:38:01.990 --> 00:38:05.290 align:middle line:84% And then imagine that this experience 00:38:05.290 --> 00:38:12.070 align:middle line:84% is so traumatizing for me and also for the fly, that I 00:38:12.070 --> 00:38:14.650 align:middle line:84% just I never complete that sentence. 00:38:14.650 --> 00:38:17.710 align:middle line:84% So I say, "This is the--" and then I stop. 00:38:17.710 --> 00:38:21.400 align:middle line:90% 00:38:21.400 --> 00:38:24.332 align:middle line:84% There are two attitudes that we could have to me 00:38:24.332 --> 00:38:25.540 align:middle line:90% having uttered that sentence. 00:38:25.540 --> 00:38:26.790 align:middle line:90% That's the sentence I uttered. 00:38:26.790 --> 00:38:30.250 align:middle line:90% "This is the cough, ugh, puh." 00:38:30.250 --> 00:38:31.630 align:middle line:90% That's something I said. 00:38:31.630 --> 00:38:35.143 align:middle line:84% And I'm a native speaker of English. 00:38:35.143 --> 00:38:37.060 align:middle line:84% So there are two kinds of things we could say. 00:38:37.060 --> 00:38:39.760 align:middle line:84% One would be to say, we're developing 00:38:39.760 --> 00:38:42.130 align:middle line:84% a theory of all of the kinds of sentences 00:38:42.130 --> 00:38:44.200 align:middle line:84% that Native English speakers can say. 00:38:44.200 --> 00:38:46.150 align:middle line:90% And that was one. 00:38:46.150 --> 00:38:48.910 align:middle line:84% "This is the cough hack splutter." 00:38:48.910 --> 00:38:51.310 align:middle line:84% And so we want a theory that allows 00:38:51.310 --> 00:38:52.960 align:middle line:84% that to be a sentence of English. 00:38:52.960 --> 00:38:55.890 align:middle line:84% That's one kind of thing we could say. 00:38:55.890 --> 00:38:58.500 align:middle line:84% This is one of those moments that happens a lot in classes 00:38:58.500 --> 00:38:59.880 align:middle line:84% where the professor says, "Here's 00:38:59.880 --> 00:39:02.520 align:middle line:84% one thing we could say," and then describes something 00:39:02.520 --> 00:39:04.500 align:middle line:90% completely ridiculous. 00:39:04.500 --> 00:39:06.157 align:middle line:90% You could do that. 00:39:06.157 --> 00:39:07.740 align:middle line:84% But here's another thing we could say. 00:39:07.740 --> 00:39:10.355 align:middle line:84% We could say, no, look, if we're developing a theory of all 00:39:10.355 --> 00:39:12.480 align:middle line:84% of the things that native English speakers can say, 00:39:12.480 --> 00:39:14.105 align:middle line:84% native speakers of any language at all, 00:39:14.105 --> 00:39:16.860 align:middle line:84% but we're going to start with English, 00:39:16.860 --> 00:39:19.830 align:middle line:84% what we want is not a theory that covers "This 00:39:19.830 --> 00:39:21.218 align:middle line:90% is the cough, hack, splutter." 00:39:21.218 --> 00:39:23.010 align:middle line:84% That's not going to be one of the sentences 00:39:23.010 --> 00:39:24.093 align:middle line:90% we're going to try to get. 00:39:24.093 --> 00:39:26.280 align:middle line:84% We're going to try to get sentences like, "This 00:39:26.280 --> 00:39:30.510 align:middle line:90% is the answer" or whatever. 00:39:30.510 --> 00:39:34.350 align:middle line:84% That's going to be a sentence, whatever it is I meant to say. 00:39:34.350 --> 00:39:36.660 align:middle line:84% And then there are going to be other things. 00:39:36.660 --> 00:39:43.350 align:middle line:84% Flies, sudden heart attacks, and then 00:39:43.350 --> 00:39:45.458 align:middle line:84% other kinds of things that are maybe 00:39:45.458 --> 00:39:47.250 align:middle line:84% less clear to think what to say about them. 00:39:47.250 --> 00:39:49.020 align:middle line:90% Flies, sure. 00:39:49.020 --> 00:39:50.615 align:middle line:90% Sudden heart attacks, yeah. 00:39:50.615 --> 00:39:51.990 align:middle line:84% What about if I forget what I was 00:39:51.990 --> 00:39:53.790 align:middle line:84% going to say in the middle of a sentence. 00:39:53.790 --> 00:39:55.380 align:middle line:84% I know that's hard to imagine, but imagine 00:39:55.380 --> 00:39:56.580 align:middle line:90% that I did something like that. 00:39:56.580 --> 00:39:58.622 align:middle line:84% I'm in the middle of a sentence, and I'm talking, 00:39:58.622 --> 00:40:01.863 align:middle line:84% and then I just forget where I was going, and-- 00:40:01.863 --> 00:40:03.030 align:middle line:90% where was I going with that? 00:40:03.030 --> 00:40:04.980 align:middle line:90% Who knows. 00:40:04.980 --> 00:40:06.580 align:middle line:90% That could happen. 00:40:06.580 --> 00:40:08.290 align:middle line:90% So what we're going to have-- 00:40:08.290 --> 00:40:10.800 align:middle line:84% so this is a different approach, and it's the one 00:40:10.800 --> 00:40:12.630 align:middle line:84% that you might imagine I'm recommending. 00:40:12.630 --> 00:40:14.130 align:middle line:84% What we're going to have is the idea 00:40:14.130 --> 00:40:16.260 align:middle line:84% that we're going to develop a theory of what 00:40:16.260 --> 00:40:18.210 align:middle line:84% English speakers say, but we're going 00:40:18.210 --> 00:40:20.610 align:middle line:84% to imagine the kind of English speaker 00:40:20.610 --> 00:40:24.090 align:middle line:84% who never inhales flies, and never forgets what they were 00:40:24.090 --> 00:40:26.640 align:middle line:84% going to say, never has a fatal heart attack, 00:40:26.640 --> 00:40:32.480 align:middle line:84% and only speaks in completely grammatical sentences. 00:40:32.480 --> 00:40:36.710 align:middle line:84% There might not be any speakers like that. 00:40:36.710 --> 00:40:39.230 align:middle line:84% If you have ever looked at a transcript of somebody 00:40:39.230 --> 00:40:44.210 align:middle line:84% talking, there aren't any sentences in there, 00:40:44.210 --> 00:40:47.510 align:middle line:84% unless the person is reading a text, 00:40:47.510 --> 00:40:50.930 align:middle line:84% or unless the person is Noam Chomsky, I have to say. 00:40:50.930 --> 00:40:53.510 align:middle line:84% It was quite weird reading transcripts 00:40:53.510 --> 00:40:56.030 align:middle line:84% of Noam Chomsky talking, or listening to Noam Chomsky talk, 00:40:56.030 --> 00:40:59.420 align:middle line:84% because, in fact, he talks in complete sentences, paragraphs. 00:40:59.420 --> 00:41:01.460 align:middle line:90% It's kind of astonishing. 00:41:01.460 --> 00:41:03.860 align:middle line:84% Normally, if you look at the transcript of somebody 00:41:03.860 --> 00:41:05.397 align:middle line:90% talking-- 00:41:05.397 --> 00:41:07.730 align:middle line:84% speaking of Noam Chomsky, I've often heard him say this. 00:41:07.730 --> 00:41:10.490 align:middle line:84% Journalists know that the best way 00:41:10.490 --> 00:41:12.920 align:middle line:84% to make someone look like a complete idiot is 00:41:12.920 --> 00:41:14.630 align:middle line:90% to quote them accurately. 00:41:14.630 --> 00:41:16.250 align:middle line:84% Just to write down exactly what they 00:41:16.250 --> 00:41:19.172 align:middle line:84% said, because they'll say "um" and "uh" and they'll stop, 00:41:19.172 --> 00:41:21.380 align:middle line:84% and they'll pause, and they'll change what they said, 00:41:21.380 --> 00:41:22.730 align:middle line:90% and they'll inhale flies. 00:41:22.730 --> 00:41:27.590 align:middle line:84% Things will happen such that their sentences are not fully 00:41:27.590 --> 00:41:29.870 align:middle line:90% grammatical sentences. 00:41:29.870 --> 00:41:32.900 align:middle line:84% What journalists, in fact, do is to clean up all that stuff 00:41:32.900 --> 00:41:34.520 align:middle line:84% so that people sounded like they were 00:41:34.520 --> 00:41:36.530 align:middle line:90% talking in complete sentences. 00:41:36.530 --> 00:41:38.570 align:middle line:84% So we're going to develop a theory of what 00:41:38.570 --> 00:41:40.100 align:middle line:84% English speakers say, but it's going 00:41:40.100 --> 00:41:44.900 align:middle line:84% to be a theory that's divorced from reality 00:41:44.900 --> 00:41:46.400 align:middle line:90% to a certain extent. 00:41:46.400 --> 00:41:49.010 align:middle line:84% We're going to imagine what people would be like 00:41:49.010 --> 00:41:51.650 align:middle line:84% if there were no distractions, and no flies, 00:41:51.650 --> 00:41:55.580 align:middle line:84% and no sudden homicides, no falling 00:41:55.580 --> 00:41:59.490 align:middle line:84% asleep in the middle of your sentences, all of that stuff. 00:41:59.490 --> 00:42:03.230 align:middle line:84% So the distinction here is competence versus performance. 00:42:03.230 --> 00:42:05.240 align:middle line:84% We're imagining a speaker who's kind 00:42:05.240 --> 00:42:10.775 align:middle line:84% of like a frictionless plane, that there 00:42:10.775 --> 00:42:12.590 align:middle line:84% are various kinds of complications, 00:42:12.590 --> 00:42:14.720 align:middle line:84% and there's no air resistance or whatever else. 00:42:14.720 --> 00:42:17.877 align:middle line:84% Various kinds of complications don't arise. 00:42:17.877 --> 00:42:20.210 align:middle line:84% So we're going to be talking about speakers' competence, 00:42:20.210 --> 00:42:23.120 align:middle line:84% what they would do if there were no distractions 00:42:23.120 --> 00:42:24.600 align:middle line:90% and no problems. 00:42:24.600 --> 00:42:26.090 align:middle line:90% There's also performance. 00:42:26.090 --> 00:42:28.670 align:middle line:84% That's the study of what people actually do. 00:42:28.670 --> 00:42:30.500 align:middle line:84% And we want to study that, sure, but we're 00:42:30.500 --> 00:42:32.810 align:middle line:84% going to develop a theory of competence on the theory 00:42:32.810 --> 00:42:35.690 align:middle line:84% that it'll be simpler in the hope 00:42:35.690 --> 00:42:39.562 align:middle line:84% that by abstracting away from various kinds of complications, 00:42:39.562 --> 00:42:41.520 align:middle line:84% we'll get a clearer picture of what's going on. 00:42:41.520 --> 00:42:42.290 align:middle line:90% Does that makes sense? 00:42:42.290 --> 00:42:43.790 align:middle line:84% That's how we're going to do syntax. 00:42:43.790 --> 00:42:48.950 align:middle line:90% 00:42:48.950 --> 00:42:53.660 align:middle line:84% Similarly, "It's raining" is a possible sentence of English. 00:42:53.660 --> 00:42:55.220 align:middle line:84% "John thinks that it's raining" is 00:42:55.220 --> 00:42:56.730 align:middle line:90% a possible sentence of English. 00:42:56.730 --> 00:42:58.730 align:middle line:84% "Mary thinks that John thinks that it's raining" 00:42:58.730 --> 00:43:00.560 align:middle line:84% is a possible sentence of English. 00:43:00.560 --> 00:43:04.130 align:middle line:84% In fact, for any sentence in English, 00:43:04.130 --> 00:43:07.640 align:middle line:84% it's always possible to create a longer sentence, 00:43:07.640 --> 00:43:09.320 align:middle line:90% so take any sentence, S. 00:43:09.320 --> 00:43:13.610 align:middle line:84% Here's a recipe for another sentence of English. 00:43:13.610 --> 00:43:21.230 align:middle line:84% You can always say, "She thinks that S," 00:43:21.230 --> 00:43:24.230 align:middle line:84% where she maybe refers to different people 00:43:24.230 --> 00:43:25.530 align:middle line:90% in every clause. 00:43:25.530 --> 00:43:27.628 align:middle line:90% So you can say, "It's raining." 00:43:27.628 --> 00:43:28.920 align:middle line:90% "She thinks that it's raining." 00:43:28.920 --> 00:43:30.510 align:middle line:84% "She thinks that she thinks that it's raining." 00:43:30.510 --> 00:43:33.093 align:middle line:84% "She thinks that she thinks that she thinks that it's raining" 00:43:33.093 --> 00:43:35.510 align:middle line:90% can keep going arbitrarily long. 00:43:35.510 --> 00:43:42.350 align:middle line:84% There is no bound on the length of English sentences. 00:43:42.350 --> 00:43:44.240 align:middle line:84% When we say it that way, you can tell 00:43:44.240 --> 00:43:48.260 align:middle line:84% that I am talking about competence because no matter 00:43:48.260 --> 00:43:50.720 align:middle line:84% how many recordings of English speakers you go through, 00:43:50.720 --> 00:43:53.780 align:middle line:84% you will never find an infinitely long sentence. 00:43:53.780 --> 00:43:56.390 align:middle line:84% Nobody actually says these things. 00:43:56.390 --> 00:44:01.130 align:middle line:84% But the reason nobody says an infinitely long sentence, 00:44:01.130 --> 00:44:02.390 align:middle line:90% the idea is going to be-- 00:44:02.390 --> 00:44:04.890 align:middle line:90% That isn't a fact about grammar. 00:44:04.890 --> 00:44:06.440 align:middle line:90% It's a fact about life. 00:44:06.440 --> 00:44:09.410 align:middle line:84% And we don't care about life in this class. 00:44:09.410 --> 00:44:12.292 align:middle line:84% The fact that if I were to start saying, 00:44:12.292 --> 00:44:14.750 align:middle line:84% "She thinks that she thinks that she thinks that she thinks 00:44:14.750 --> 00:44:16.910 align:middle line:84% that she thinks that she thinks that she thinks that she thinks 00:44:16.910 --> 00:44:19.118 align:middle line:84% that it's raining," then eventually people would stop 00:44:19.118 --> 00:44:21.380 align:middle line:84% listening to me, or I would run out of breath, 00:44:21.380 --> 00:44:24.020 align:middle line:84% or I would need to take a break to eat, or I would die-- 00:44:24.020 --> 00:44:26.660 align:middle line:84% there are various reasons that I will eventually 00:44:26.660 --> 00:44:29.870 align:middle line:84% stop uttering my infinitely long sentence, 00:44:29.870 --> 00:44:32.237 align:middle line:84% but we're not going to have a grammar that says-- 00:44:32.237 --> 00:44:34.820 align:middle line:84% we're not going to try to find out what's the longest sentence 00:44:34.820 --> 00:44:37.970 align:middle line:84% anybody ever uttered and try to get that fact 00:44:37.970 --> 00:44:40.218 align:middle line:84% to be a fact that we want our grammar of English, 00:44:40.218 --> 00:44:42.260 align:middle line:84% our theory of the possible sentences of English-- 00:44:42.260 --> 00:44:44.930 align:middle line:84% We're not going to try to predict that. 00:44:44.930 --> 00:44:46.430 align:middle line:84% What we're going to have is a theory 00:44:46.430 --> 00:44:49.670 align:middle line:84% that says English sentences can be arbitrarily long. 00:44:49.670 --> 00:44:51.733 align:middle line:84% And then, yeah, eventually people die, 00:44:51.733 --> 00:44:53.150 align:middle line:84% and so nobody ever says a sentence 00:44:53.150 --> 00:44:54.108 align:middle line:90% that's infinitely long. 00:44:54.108 --> 00:44:55.220 align:middle line:90% But that's about death. 00:44:55.220 --> 00:44:56.803 align:middle line:84% That's not about-- and we're not going 00:44:56.803 --> 00:44:59.220 align:middle line:84% to talk about death in this class, except when we do. 00:44:59.220 --> 00:45:00.270 align:middle line:90% Does that make sense? 00:45:00.270 --> 00:45:01.468 align:middle line:90% Yeah? 00:45:01.468 --> 00:45:03.260 align:middle line:84% So that's another move we're going to make, 00:45:03.260 --> 00:45:05.792 align:middle line:84% another instance of us caring about the difference 00:45:05.792 --> 00:45:07.250 align:middle line:84% between competence and performance. 00:45:07.250 --> 00:45:10.250 align:middle line:84% Nobody ever performs an infinitely long sentence, 00:45:10.250 --> 00:45:13.842 align:middle line:84% but we're competent to produce them. 00:45:13.842 --> 00:45:15.800 align:middle line:84% So enough talking about what we're going to do. 00:45:15.800 --> 00:45:17.330 align:middle line:90% Let's begin doing it. 00:45:17.330 --> 00:45:21.170 align:middle line:84% Questions before I begin doing some syntax? 00:45:21.170 --> 00:45:23.600 align:middle line:84% Here's the sentence, "I will find the red book." 00:45:23.600 --> 00:45:25.220 align:middle line:90% grammatical sentence. 00:45:25.220 --> 00:45:25.775 align:middle line:90% Acceptable. 00:45:25.775 --> 00:45:27.650 align:middle line:84% It's clear what it means, although we've just 00:45:27.650 --> 00:45:29.692 align:middle line:84% said it doesn't matter whether it means anything. 00:45:29.692 --> 00:45:31.880 align:middle line:90% 00:45:31.880 --> 00:45:35.930 align:middle line:84% I said early on we're going to want 00:45:35.930 --> 00:45:40.820 align:middle line:84% to have a way of saying which parts of this sentence 00:45:40.820 --> 00:45:45.320 align:middle line:84% were put together as units, like with "unlockable." 00:45:45.320 --> 00:45:47.780 align:middle line:84% We wanted to be able to say "unlockable" 00:45:47.780 --> 00:45:50.870 align:middle line:84% is ambiguous because it can consist of a unit "unlock" 00:45:50.870 --> 00:45:53.513 align:middle line:84% to which you've added "-able," or a unit "lockable" 00:45:53.513 --> 00:45:54.680 align:middle line:90% to which you've added "un-." 00:45:54.680 --> 00:45:57.003 align:middle line:84% That's what that ambiguity comes from. 00:45:57.003 --> 00:45:59.420 align:middle line:84% And we're going to do a similar kind of thing with syntax. 00:45:59.420 --> 00:46:01.180 align:middle line:84% We're going to look for these units. 00:46:01.180 --> 00:46:02.930 align:middle line:84% And what we're going to find is that there 00:46:02.930 --> 00:46:05.960 align:middle line:84% are various phenomena that care about 00:46:05.960 --> 00:46:09.440 align:middle line:84% whether something is a unit in that tree, a single blob 00:46:09.440 --> 00:46:11.580 align:middle line:90% of structure. 00:46:11.580 --> 00:46:14.250 align:middle line:84% So in a sentence like, "I will find the red book" for example, 00:46:14.250 --> 00:46:17.070 align:middle line:84% we'll see that syntax treats that string, 00:46:17.070 --> 00:46:20.130 align:middle line:90% "The red book" as a unit. 00:46:20.130 --> 00:46:22.560 align:middle line:84% There are various phenomena that care about that. 00:46:22.560 --> 00:46:23.990 align:middle line:84% So one of them is what's sometimes 00:46:23.990 --> 00:46:25.340 align:middle line:90% called topicalization. 00:46:25.340 --> 00:46:30.920 align:middle line:84% It's possible to say things like "The red book I will find." 00:46:30.920 --> 00:46:33.050 align:middle line:84% For me at least, it's easiest to say things 00:46:33.050 --> 00:46:35.150 align:middle line:84% like that if I follow it up with "The blue book, 00:46:35.150 --> 00:46:37.820 align:middle line:90% I will leave right where it is." 00:46:37.820 --> 00:46:42.560 align:middle line:84% It's OK to say things like "The red book I will find." 00:46:42.560 --> 00:46:46.670 align:middle line:84% It's OK to take a substring like that, "the red book" 00:46:46.670 --> 00:46:51.390 align:middle line:84% and put it together with another similar substring conjoined 00:46:51.390 --> 00:46:53.390 align:middle line:84% with the word "and." so you can say things like, 00:46:53.390 --> 00:46:56.990 align:middle line:84% "I will find the red book and the blue pencils." 00:46:56.990 --> 00:47:03.060 align:middle line:84% It is OK to use "the red book" as a possible answer 00:47:03.060 --> 00:47:04.180 align:middle line:90% to a question. 00:47:04.180 --> 00:47:08.070 align:middle line:84% This is like the stuff we were saying before about things you 00:47:08.070 --> 00:47:10.620 align:middle line:90% can say if you're astonished. 00:47:10.620 --> 00:47:15.420 align:middle line:84% So you can say "the red book" as, basically, a sentence 00:47:15.420 --> 00:47:16.720 align:middle line:90% under the right circumstances. 00:47:16.720 --> 00:47:18.928 align:middle line:84% For example, if somebody has just asked you what will 00:47:18.928 --> 00:47:20.580 align:middle line:90% you find. 00:47:20.580 --> 00:47:22.650 align:middle line:84% These are all ways in which "the red book" 00:47:22.650 --> 00:47:26.790 align:middle line:84% is treated as a single object that syntax gets to refer to. 00:47:26.790 --> 00:47:29.370 align:middle line:84% I hope I was smart enough to contrast it with something 00:47:29.370 --> 00:47:30.850 align:middle line:90% else on the next slide. 00:47:30.850 --> 00:47:31.350 align:middle line:90% Yes, I was. 00:47:31.350 --> 00:47:34.890 align:middle line:84% But let me give you one other test. 00:47:34.890 --> 00:47:37.710 align:middle line:90% Yeah, this is a test. 00:47:37.710 --> 00:47:40.780 align:middle line:84% I can say things like "what I will find is the red book." 00:47:40.780 --> 00:47:43.260 align:middle line:84% So I can rearrange the words of the sentence in a way such 00:47:43.260 --> 00:47:46.000 align:middle line:84% that there is a word, "is" before that string 00:47:46.000 --> 00:47:46.950 align:middle line:90% "the red book." 00:47:46.950 --> 00:47:50.650 align:middle line:84% It puts a special kind of emphasis on "the red book." 00:47:50.650 --> 00:47:52.470 align:middle line:90% It's called clefting. 00:47:52.470 --> 00:47:54.690 align:middle line:84% Contrast that with-- so this is not just 00:47:54.690 --> 00:47:57.880 align:middle line:84% a property of every three-word string in the sentence. 00:47:57.880 --> 00:48:01.560 align:middle line:84% So "find the red," for example, is not a constituent. 00:48:01.560 --> 00:48:02.400 align:middle line:90% It's not a phrase. 00:48:02.400 --> 00:48:04.400 align:middle line:84% It's not something we need to make reference to. 00:48:04.400 --> 00:48:08.310 align:middle line:84% So you cannot say things like "find the red I will book." 00:48:08.310 --> 00:48:10.350 align:middle line:84% So "the red book, I will find," "the blue book, 00:48:10.350 --> 00:48:11.700 align:middle line:90% I will leave where it is," fine. 00:48:11.700 --> 00:48:17.790 align:middle line:84% But "find the red I will book, leave the blue I will pencils." 00:48:17.790 --> 00:48:19.050 align:middle line:90% No. 00:48:19.050 --> 00:48:23.280 align:middle line:84% Can't do this with just any random three-word string. 00:48:23.280 --> 00:48:25.500 align:middle line:84% Similarly, there is no question-- 00:48:25.500 --> 00:48:27.990 align:middle line:84% there are questions to which the answer is "The red book," 00:48:27.990 --> 00:48:29.250 align:middle line:90% like "What will you find?" 00:48:29.250 --> 00:48:31.290 align:middle line:84% There are no questions to which the answer is 00:48:31.290 --> 00:48:36.050 align:middle line:84% "Find the red," apart from "What are 00:48:36.050 --> 00:48:39.590 align:middle line:84% the third, fourth, and fifth words of this sentence?" 00:48:39.590 --> 00:48:43.640 align:middle line:84% So put aside mental linguistic games like that. 00:48:43.640 --> 00:48:46.730 align:middle line:84% You can't-- and there's no question that will give you 00:48:46.730 --> 00:48:47.420 align:middle line:90% that answer. 00:48:47.420 --> 00:48:48.140 align:middle line:90% Yeah? 00:48:48.140 --> 00:48:50.090 align:middle line:84% AUDIENCE: What color do I need to find? 00:48:50.090 --> 00:48:52.090 align:middle line:84% NORVIN RICHARDS: "What color do I need to find?" 00:48:52.090 --> 00:48:54.000 align:middle line:90% "Find the red." 00:48:54.000 --> 00:48:55.540 align:middle line:90% Really? 00:48:55.540 --> 00:48:56.470 align:middle line:90% Oh, I see. 00:48:56.470 --> 00:49:00.210 align:middle line:84% You mean "What color book do I need to find?" 00:49:00.210 --> 00:49:02.390 align:middle line:90% "Find the red." 00:49:02.390 --> 00:49:04.790 align:middle line:84% AUDIENCE: How [? about ?] "Find the red one?" 00:49:04.790 --> 00:49:05.390 align:middle line:90% NORVIN RICHARDS: Find the red-- 00:49:05.390 --> 00:49:06.785 align:middle line:84% I want to say "Find the red one." 00:49:06.785 --> 00:49:09.640 align:middle line:90% 00:49:09.640 --> 00:49:10.140 align:middle line:90% Yes. 00:49:10.140 --> 00:49:13.640 align:middle line:90% 00:49:13.640 --> 00:49:18.190 align:middle line:84% Is there a faster way to convince you of that? 00:49:18.190 --> 00:49:20.950 align:middle line:84% Or similarly, if I switch to my other way 00:49:20.950 --> 00:49:25.000 align:middle line:84% of eliciting fragments of sentences, if I tell you 00:49:25.000 --> 00:49:27.700 align:middle line:84% I will find the red book and you're amazed, 00:49:27.700 --> 00:49:29.770 align:middle line:90% you can say, "The red book?" 00:49:29.770 --> 00:49:32.380 align:middle line:84% But if I tell you that I will find the red book 00:49:32.380 --> 00:49:35.890 align:middle line:84% and you're amazed, you're not going to say, "Find the red?" 00:49:35.890 --> 00:49:36.550 align:middle line:90% I think. 00:49:36.550 --> 00:49:38.830 align:middle line:90% Do you think that's true? 00:49:38.830 --> 00:49:41.380 align:middle line:84% But yeah, I take your point about-- 00:49:41.380 --> 00:49:43.840 align:middle line:84% to the extent that you can use "the red" as shorthand 00:49:43.840 --> 00:49:45.268 align:middle line:90% for "the red one." 00:49:45.268 --> 00:49:47.560 align:middle line:84% AUDIENCE: It's red acting as the noun, not red acting-- 00:49:47.560 --> 00:49:48.190 align:middle line:90% NORVIN RICHARDS: Yeah. 00:49:48.190 --> 00:49:48.580 align:middle line:90% Yeah. 00:49:48.580 --> 00:49:48.940 align:middle line:90% Yeah, no. 00:49:48.940 --> 00:49:50.170 align:middle line:90% That's an interesting point. 00:49:50.170 --> 00:49:52.003 align:middle line:84% Several people have points about that point. 00:49:52.003 --> 00:49:53.177 align:middle line:90% Joseph? 00:49:53.177 --> 00:49:55.510 align:middle line:84% AUDIENCE: I think "Find the red" is an acceptable answer 00:49:55.510 --> 00:49:57.450 align:middle line:84% to a question that you're finding this-- 00:49:57.450 --> 00:49:59.700 align:middle line:84% suppose you have this fictional-- 00:49:59.700 --> 00:50:01.200 align:middle line:84% this children's game where there's 00:50:01.200 --> 00:50:02.550 align:middle line:90% a bunch of little tiles. 00:50:02.550 --> 00:50:04.125 align:middle line:90% You have to find the red-- 00:50:04.125 --> 00:50:05.250 align:middle line:90% NORVIN RICHARDS: Yeah yeah. 00:50:05.250 --> 00:50:06.270 align:middle line:90% AUDIENCE: --the red ones. 00:50:06.270 --> 00:50:06.540 align:middle line:90% NORVIN RICHARDS: Yeah. 00:50:06.540 --> 00:50:08.003 align:middle line:84% So this is like your example where 00:50:08.003 --> 00:50:09.420 align:middle line:84% we're going to treat red as a noun 00:50:09.420 --> 00:50:10.628 align:middle line:90% that you're going to go find. 00:50:10.628 --> 00:50:12.712 align:middle line:84% AUDIENCE: I guess this is kind of the same thing-- 00:50:12.712 --> 00:50:14.400 align:middle line:84% "land of the free," "home of the brave." 00:50:14.400 --> 00:50:15.180 align:middle line:90% NORVIN RICHARDS: Oh. 00:50:15.180 --> 00:50:15.680 align:middle line:90% Yeah. 00:50:15.680 --> 00:50:19.170 align:middle line:84% So we have some cases where we have things 00:50:19.170 --> 00:50:21.300 align:middle line:84% that you certainly-- should be adjectives, 00:50:21.300 --> 00:50:23.970 align:middle line:84% that either we're getting to use them as nouns, 00:50:23.970 --> 00:50:28.272 align:middle line:84% or we're getting to modify nouns that you can't hear, 00:50:28.272 --> 00:50:29.730 align:middle line:84% however we want to talk about that. 00:50:29.730 --> 00:50:30.438 align:middle line:90% Yeah, good point. 00:50:30.438 --> 00:50:31.950 align:middle line:90% Good point. 00:50:31.950 --> 00:50:32.630 align:middle line:90% Yeah? 00:50:32.630 --> 00:50:33.630 align:middle line:90% Other points about this? 00:50:33.630 --> 00:50:36.330 align:middle line:90% 00:50:36.330 --> 00:50:38.400 align:middle line:84% So all this slide is meant to convince 00:50:38.400 --> 00:50:40.845 align:middle line:84% you of is that "the red book" and "find the red" 00:50:40.845 --> 00:50:41.970 align:middle line:90% don't have the same status. 00:50:41.970 --> 00:50:44.340 align:middle line:84% "The red book," we want it to be a substring that 00:50:44.340 --> 00:50:47.910 align:middle line:84% has certain privileges, can be used for these various types 00:50:47.910 --> 00:50:51.070 align:middle line:84% of phenomena, as opposed to "find the red," 00:50:51.070 --> 00:50:53.100 align:middle line:84% which you can't do those things with. 00:50:53.100 --> 00:50:55.950 align:middle line:84% So it isn't just these are phenomena that pick out 00:50:55.950 --> 00:50:58.117 align:middle line:90% three-words substrings. 00:50:58.117 --> 00:50:59.700 align:middle line:84% It's these are phenomena that pick out 00:50:59.700 --> 00:51:01.232 align:middle line:84% certain substrings and not others 00:51:01.232 --> 00:51:02.940 align:middle line:84% of the sentence, certain strings of words 00:51:02.940 --> 00:51:05.550 align:middle line:84% and not others of the sentence, and we're 00:51:05.550 --> 00:51:08.275 align:middle line:84% going to try to figure out theories about why 00:51:08.275 --> 00:51:09.525 align:middle line:90% those are the special strings. 00:51:09.525 --> 00:51:15.800 align:middle line:90% 00:51:15.800 --> 00:51:19.310 align:middle line:84% And what we're going to do is we'll say just 00:51:19.310 --> 00:51:23.840 align:middle line:84% as with "unlockable" we were taking pairs of things 00:51:23.840 --> 00:51:27.560 align:middle line:84% and putting them together to form larger things, 00:51:27.560 --> 00:51:28.730 align:middle line:90% larger units. 00:51:28.730 --> 00:51:32.120 align:middle line:84% We'll do the same thing here, only with words. 00:51:32.120 --> 00:51:35.120 align:middle line:84% We had this operation, we were calling it merge, 00:51:35.120 --> 00:51:37.250 align:middle line:84% that takes pairs of things and puts them together 00:51:37.250 --> 00:51:38.510 align:middle line:90% into a larger thing. 00:51:38.510 --> 00:51:40.400 align:middle line:84% And similarly here, what we're going 00:51:40.400 --> 00:51:43.730 align:middle line:84% to do to create a sentence like "I will find the red book," 00:51:43.730 --> 00:51:45.590 align:middle line:84% we'll start with just the end of it, 00:51:45.590 --> 00:51:47.780 align:middle line:84% "find the red book" is we're going 00:51:47.780 --> 00:51:49.820 align:middle line:84% to take pairs of things like "red" and "book" 00:51:49.820 --> 00:51:51.020 align:middle line:90% and put them together. 00:51:51.020 --> 00:51:53.237 align:middle line:84% And then we'll take that unit that we've 00:51:53.237 --> 00:51:55.070 align:middle line:84% created by putting together "red" and "book" 00:51:55.070 --> 00:51:58.160 align:middle line:84% and we'll put that together with this word, "the." 00:51:58.160 --> 00:52:02.420 align:middle line:84% And then we'll take that thing that we've 00:52:02.420 --> 00:52:07.850 align:middle line:84% created by putting together "the red book" and we'll add "find." 00:52:07.850 --> 00:52:10.820 align:middle line:84% So just as with "unlockable," we were taking pairs of things 00:52:10.820 --> 00:52:14.435 align:middle line:84% and putting them together in pairs 00:52:14.435 --> 00:52:17.000 align:middle line:84% to create these larger and larger structures. 00:52:17.000 --> 00:52:20.565 align:middle line:84% We're going to do the same thing to create sentences out 00:52:20.565 --> 00:52:21.065 align:middle line:90% of words. 00:52:21.065 --> 00:52:24.560 align:middle line:90% 00:52:24.560 --> 00:52:27.200 align:middle line:84% This way of talking about it has the virtue 00:52:27.200 --> 00:52:28.910 align:middle line:84% of giving us a vocabulary for talking 00:52:28.910 --> 00:52:30.910 align:middle line:84% about those kinds of observations we were making 00:52:30.910 --> 00:52:32.880 align:middle line:90% on the last two slides. 00:52:32.880 --> 00:52:38.042 align:middle line:84% So when we say "the red book" is a unit that various things get 00:52:38.042 --> 00:52:40.250 align:middle line:84% to apply to, things like what I called topicalization 00:52:40.250 --> 00:52:42.890 align:middle line:84% where you take a chunk of the sentence 00:52:42.890 --> 00:52:46.822 align:middle line:84% and put it at the beginning, and it has some kind of emphasis. 00:52:46.822 --> 00:52:48.530 align:middle line:84% Because this is syntax and not semantics, 00:52:48.530 --> 00:52:52.280 align:middle line:84% we won't worry too much about what it means. 00:52:52.280 --> 00:52:54.170 align:middle line:84% When we say that's something you can 00:52:54.170 --> 00:52:56.960 align:middle line:84% do to the string "the red book," but not, for example, 00:52:56.960 --> 00:52:58.940 align:middle line:84% to "find the red," this tree gives us 00:52:58.940 --> 00:53:00.140 align:middle line:90% a way of talking about that. 00:53:00.140 --> 00:53:03.830 align:middle line:84% There is a unit that we created in the course of putting things 00:53:03.830 --> 00:53:06.980 align:middle line:84% together in pairs that is just "the red book." 00:53:06.980 --> 00:53:09.800 align:middle line:84% it's the unit that I've circled there. 00:53:09.800 --> 00:53:12.560 align:middle line:84% But there is no unit that we've created 00:53:12.560 --> 00:53:15.050 align:middle line:84% as we've been putting these pairs together that consists 00:53:15.050 --> 00:53:18.430 align:middle line:90% just to "find the red." 00:53:18.430 --> 00:53:20.540 align:middle line:90% Do people see that in this tree? 00:53:20.540 --> 00:53:23.590 align:middle line:84% So there's a node in the tree, if you want. 00:53:23.590 --> 00:53:28.330 align:middle line:84% It's the one that I circled in red that consists just 00:53:28.330 --> 00:53:30.760 align:middle line:90% of the words "the red book." 00:53:30.760 --> 00:53:32.830 align:middle line:84% But there is no thing that I could 00:53:32.830 --> 00:53:35.815 align:middle line:84% circle that would consist just of the words, "find the red." 00:53:35.815 --> 00:53:38.137 align:middle line:90% 00:53:38.137 --> 00:53:39.720 align:middle line:84% There are other things I could circle. 00:53:39.720 --> 00:53:41.580 align:middle line:84% I could have circled the largest thing, 00:53:41.580 --> 00:53:44.280 align:middle line:84% but that corresponds to "find the red book." 00:53:44.280 --> 00:53:47.820 align:middle line:84% Or I could have circled a thing that consists just 00:53:47.820 --> 00:53:48.780 align:middle line:90% of "red" and "book." 00:53:48.780 --> 00:53:50.160 align:middle line:90% It would just be "red book." 00:53:50.160 --> 00:53:52.800 align:middle line:84% But there's nothing that's just "find the red." 00:53:52.800 --> 00:53:54.750 align:middle line:84% And that's what we're going to relate 00:53:54.750 --> 00:53:57.000 align:middle line:84% to all those observations we made on those two slides. 00:53:57.000 --> 00:54:05.340 align:middle line:90% 00:54:05.340 --> 00:54:08.027 align:middle line:84% I've said this now a couple of times, 00:54:08.027 --> 00:54:10.110 align:middle line:84% this is meant to remind you of stuff we were doing 00:54:10.110 --> 00:54:11.190 align:middle line:90% when we were doing morphology. 00:54:11.190 --> 00:54:12.565 align:middle line:84% When we were doing morphology, we 00:54:12.565 --> 00:54:15.090 align:middle line:84% were using this operation, we called it merge, 00:54:15.090 --> 00:54:19.650 align:middle line:84% that assembled pairs of things and created new things. 00:54:19.650 --> 00:54:21.630 align:middle line:84% When we were doing morphology, we 00:54:21.630 --> 00:54:24.100 align:middle line:84% were then saying when you put these two things together, 00:54:24.100 --> 00:54:26.910 align:middle line:84% you have a new thing, and we're giving that thing 00:54:26.910 --> 00:54:31.140 align:middle line:84% a label whose properties are determined by the things 00:54:31.140 --> 00:54:32.950 align:middle line:90% that you put together. 00:54:32.950 --> 00:54:37.128 align:middle line:84% So when you put together "un-" number one, and "lock," 00:54:37.128 --> 00:54:38.670 align:middle line:84% it's part of the specification of un- 00:54:38.670 --> 00:54:40.670 align:middle line:84% number one that when you combine it with a verb, 00:54:40.670 --> 00:54:42.510 align:middle line:90% what you get is a verb. 00:54:42.510 --> 00:54:47.190 align:middle line:84% So the tree on the left there, "lock" is labeled as a verb, 00:54:47.190 --> 00:54:49.650 align:middle line:84% and "unlock" is also labeled as a verb. 00:54:49.650 --> 00:54:51.780 align:middle line:90% People see that. 00:54:51.780 --> 00:54:54.910 align:middle line:84% And then when you combine that verb "unlock" with a "-able," 00:54:54.910 --> 00:54:58.560 align:middle line:84% it's a property of "-able" that it merges with verbs 00:54:58.560 --> 00:55:01.507 align:middle line:84% and the thing that you create as a result is an adjective. 00:55:01.507 --> 00:55:03.090 align:middle line:84% And so the whole thing is "unlockable. 00:55:03.090 --> 00:55:05.958 align:middle line:84% This is what we were doing with labels before. 00:55:05.958 --> 00:55:08.250 align:middle line:84% We're going to want to do something similar for syntax. 00:55:08.250 --> 00:55:10.500 align:middle line:84% We're going to want to label the parts of these trees. 00:55:10.500 --> 00:55:13.312 align:middle line:90% 00:55:13.312 --> 00:55:15.020 align:middle line:84% What kinds of labels are we going to use? 00:55:15.020 --> 00:55:18.352 align:middle line:84% Well, look, I just gave you all these diagnostics 00:55:18.352 --> 00:55:20.560 align:middle line:84% to try to convince you that when you say things like, 00:55:20.560 --> 00:55:23.080 align:middle line:84% "I will find the red book," "the red book" 00:55:23.080 --> 00:55:29.050 align:middle line:84% should be a unit, right, phenomena like topicalization 00:55:29.050 --> 00:55:32.188 align:middle line:84% should get to make reference to it. 00:55:32.188 --> 00:55:33.730 align:middle line:84% We have to decide what to label that. 00:55:33.730 --> 00:55:37.570 align:middle line:84% Notice there are a bunch of things that 00:55:37.570 --> 00:55:38.960 align:middle line:90% have all of those properties. 00:55:38.960 --> 00:55:42.310 align:middle line:84% So if I said, "I will find red books," well, 00:55:42.310 --> 00:55:44.620 align:middle line:84% you'd be able to topicalized "red books." 00:55:44.620 --> 00:55:46.480 align:middle line:90% So "I will find red books." 00:55:46.480 --> 00:55:48.130 align:middle line:90% Or "red books I will find." 00:55:48.130 --> 00:55:52.060 align:middle line:84% "I will find red books," and you're amazed: "Red books?" 00:55:52.060 --> 00:55:54.655 align:middle line:84% Similarly, "I will find those red books about linguistics." 00:55:54.655 --> 00:55:56.530 align:middle line:84% Yeah, "those red books about linguistics," 00:55:56.530 --> 00:56:00.610 align:middle line:90% that's a unit a similar kind. 00:56:00.610 --> 00:56:03.280 align:middle line:84% You can say "Those red books about linguistics, 00:56:03.280 --> 00:56:03.940 align:middle line:90% I will find." 00:56:03.940 --> 00:56:07.570 align:middle line:84% Yes, so that's a unit we want to be able to make reference to. 00:56:07.570 --> 00:56:10.080 align:middle line:84% "I will find books" similarly "books I will find" 00:56:10.080 --> 00:56:11.230 align:middle line:90% or "I will find books." 00:56:11.230 --> 00:56:11.830 align:middle line:90% "Books?" 00:56:11.830 --> 00:56:12.850 align:middle line:90% You're amazed. 00:56:12.850 --> 00:56:15.100 align:middle line:84% So "books" by itself is apparently 00:56:15.100 --> 00:56:17.350 align:middle line:90% a unit of the same kind. 00:56:17.350 --> 00:56:22.345 align:middle line:84% So apparently the important part of that, 00:56:22.345 --> 00:56:24.520 align:middle line:84% the thing that all of those have in common, 00:56:24.520 --> 00:56:27.290 align:middle line:90% is that they contain a noun. 00:56:27.290 --> 00:56:31.710 align:middle line:84% So we're going to name that phrase after the fact 00:56:31.710 --> 00:56:33.490 align:middle line:90% that it contains a noun. 00:56:33.490 --> 00:56:37.410 align:middle line:84% So when we put together "red" and "book," what we get 00:56:37.410 --> 00:56:40.350 align:middle line:84% has properties that are determined by the fact 00:56:40.350 --> 00:56:42.550 align:middle line:90% that it contains a noun. 00:56:42.550 --> 00:56:45.030 align:middle line:84% If there were no noun, it wouldn't have those properties. 00:56:45.030 --> 00:56:46.650 align:middle line:90% Similarly, with "the red book." 00:56:46.650 --> 00:56:49.230 align:middle line:90% 00:56:49.230 --> 00:56:52.470 align:middle line:84% The things that can go in that slot, the things that 00:56:52.470 --> 00:56:56.010 align:middle line:84% can go after "find," we just saw in the last slide, there 00:56:56.010 --> 00:56:58.810 align:middle line:84% are various kinds of things that can be larger or smaller. 00:56:58.810 --> 00:57:00.610 align:middle line:90% What they all contain is a noun. 00:57:00.610 --> 00:57:02.430 align:middle line:84% So we're going to name that thing 00:57:02.430 --> 00:57:05.430 align:middle line:84% after those kinds of units, after the fact that they 00:57:05.430 --> 00:57:06.090 align:middle line:90% contain a noun. 00:57:06.090 --> 00:57:07.590 align:middle line:90% We'll give it the label "noun." 00:57:07.590 --> 00:57:10.370 align:middle line:90% 00:57:10.370 --> 00:57:13.167 align:middle line:84% Kind of like when we added "un-" to "lock" and got "unlock," 00:57:13.167 --> 00:57:14.750 align:middle line:84% we said, oh yeah, this is also a verb. 00:57:14.750 --> 00:57:17.720 align:middle line:84% It acts like a verb in every other way. 00:57:17.720 --> 00:57:19.970 align:middle line:84% You can use it as a verb: "I unlock the door." 00:57:19.970 --> 00:57:23.060 align:middle line:84% You can add things to it that can be added to verbs, 00:57:23.060 --> 00:57:24.170 align:middle line:90% like "-able." 00:57:24.170 --> 00:57:26.060 align:middle line:84% So similarly here, when we're trying 00:57:26.060 --> 00:57:29.032 align:middle line:84% to decide what kinds of things can go in this slot, 00:57:29.032 --> 00:57:31.490 align:middle line:84% the slot that's right after "find," the thing that they all 00:57:31.490 --> 00:57:33.610 align:middle line:84% have in common is that they contain nouns. 00:57:33.610 --> 00:57:35.360 align:middle line:84% So we're going to name them after the fact 00:57:35.360 --> 00:57:36.360 align:middle line:90% that they contain nouns. 00:57:36.360 --> 00:57:38.720 align:middle line:84% We'll call them all "extended nouns." 00:57:38.720 --> 00:57:39.220 align:middle line:90% Yeah? 00:57:39.220 --> 00:57:42.840 align:middle line:90% 00:57:42.840 --> 00:57:48.600 align:middle line:84% Similarly, there is another unit in "I will find the red book," 00:57:48.600 --> 00:57:52.620 align:middle line:84% not just "the red book," which we said is this extended noun. 00:57:52.620 --> 00:57:57.340 align:middle line:84% There's also another phrase, "find the red book," 00:57:57.340 --> 00:57:58.560 align:middle line:90% which is also a unit. 00:57:58.560 --> 00:58:01.978 align:middle line:84% And it was the unit that we were constructing in the slides 00:58:01.978 --> 00:58:04.020 align:middle line:84% when I was showing you trees for how to construct 00:58:04.020 --> 00:58:05.130 align:middle line:90% "find the red book." 00:58:05.130 --> 00:58:10.680 align:middle line:84% and it passes these tests for unithood, constituency. 00:58:10.680 --> 00:58:15.630 align:middle line:84% You can say, "Find the red book I will," 00:58:15.630 --> 00:58:20.130 align:middle line:84% not only if you are Yoda, but also if you say something like, 00:58:20.130 --> 00:58:21.900 align:middle line:84% "I said that I would find the red book, 00:58:21.900 --> 00:58:23.850 align:middle line:90% and find the red book I will." 00:58:23.850 --> 00:58:27.840 align:middle line:84% It's more or less OK sentences of English. 00:58:27.840 --> 00:58:31.810 align:middle line:84% Or if I say "I will find the red book" and you are amazed, 00:58:31.810 --> 00:58:33.510 align:middle line:84% you can say, "Find the red book?" 00:58:33.510 --> 00:58:35.340 align:middle line:84% (No one has ever found the red book. 00:58:35.340 --> 00:58:40.440 align:middle line:84% It's been lost for centuries.) And various other things that 00:58:40.440 --> 00:58:42.150 align:middle line:90% are on this slide. 00:58:42.150 --> 00:58:44.778 align:middle line:84% "Find the red book" is also a constituent, also a unit 00:58:44.778 --> 00:58:46.320 align:middle line:84% that we're going to want syntax to be 00:58:46.320 --> 00:58:48.330 align:middle line:90% able to make reference to. 00:58:48.330 --> 00:58:53.280 align:middle line:84% And just as we said "the red book," the important part 00:58:53.280 --> 00:58:55.500 align:middle line:84% of "the red book" is the noun, that's 00:58:55.500 --> 00:58:57.990 align:middle line:84% the thing that determines that that phrase is 00:58:57.990 --> 00:58:59.397 align:middle line:90% OK in that place. 00:58:59.397 --> 00:59:01.980 align:middle line:84% Yeah, there are various phrases that you can put in that place 00:59:01.980 --> 00:59:03.810 align:middle line:90% and they all have nouns in them. 00:59:03.810 --> 00:59:06.510 align:middle line:84% Similarly, all the things you can put in this place 00:59:06.510 --> 00:59:08.407 align:middle line:90% have verbs in them. 00:59:08.407 --> 00:59:09.990 align:middle line:84% They don't have to have anything else. 00:59:09.990 --> 00:59:12.240 align:middle line:84% You can say things like "I will leave," 00:59:12.240 --> 00:59:15.540 align:middle line:84% and then "leave" has all the properties we just ran through. 00:59:15.540 --> 00:59:18.120 align:middle line:84% "I said I would leave, and leave I will." 00:59:18.120 --> 00:59:21.240 align:middle line:84% "Leave" is a unit of the same kind as "find the red book." 00:59:21.240 --> 00:59:24.120 align:middle line:84% Or "I will leave" and you're amazed. 00:59:24.120 --> 00:59:26.010 align:middle line:90% You can say, "Leave?" 00:59:26.010 --> 00:59:28.872 align:middle line:84% So "leave" is a unit of the same kind as "find the red book." 00:59:28.872 --> 00:59:30.330 align:middle line:84% So the important part for this part 00:59:30.330 --> 00:59:34.410 align:middle line:84% is the verb, the part that determines 00:59:34.410 --> 00:59:37.180 align:middle line:84% that that's the kind of phrase that can go in that position. 00:59:37.180 --> 00:59:39.390 align:middle line:84% And so when we're putting together 00:59:39.390 --> 00:59:42.660 align:middle line:84% "find" with "the red book" we said "the red book" 00:59:42.660 --> 00:59:44.340 align:middle line:90% is kind of an extended noun. 00:59:44.340 --> 00:59:46.770 align:middle line:84% It's a large unit whose special property 00:59:46.770 --> 00:59:48.750 align:middle line:90% is that it contains a noun. 00:59:48.750 --> 00:59:52.530 align:middle line:84% "Find the red book" we're going to give that the label verb, 00:59:52.530 --> 00:59:55.870 align:middle line:84% because having a verb is the important part for that. 00:59:55.870 --> 00:59:58.110 align:middle line:90% So yeah, so far so good? 00:59:58.110 --> 01:00:01.290 align:middle line:90% 01:00:01.290 --> 01:00:06.000 align:middle line:90% So yeah, one more. 01:00:06.000 --> 01:00:08.430 align:middle line:84% You can say "I will find the book in the garage." 01:00:08.430 --> 01:00:12.330 align:middle line:84% "In the garage" is a unit, it's a constituent. 01:00:12.330 --> 01:00:14.250 align:middle line:84% It's a phrase, it's the kind of thing syntax 01:00:14.250 --> 01:00:16.930 align:middle line:90% gets to care about. 01:00:16.930 --> 01:00:19.545 align:middle line:84% And again, if I say, "I will find the book in the garage" 01:00:19.545 --> 01:00:23.520 align:middle line:84% and you're amazed, you can say "In the garage?" 01:00:23.520 --> 01:00:25.680 align:middle line:84% If I want to, I can topicalize "in the garage." 01:00:25.680 --> 01:00:29.010 align:middle line:84% I can say things like, "in the garage I will find the book." 01:00:29.010 --> 01:00:32.490 align:middle line:90% 01:00:32.490 --> 01:00:33.990 align:middle line:90% It's a prepositional phrase. 01:00:33.990 --> 01:00:39.510 align:middle line:84% The kinds of things that can go in a prepositional phrase 01:00:39.510 --> 01:00:42.930 align:middle line:84% include prepositions and noun phrases. 01:00:42.930 --> 01:00:44.880 align:middle line:84% "The garage" is another noun phrase. 01:00:44.880 --> 01:00:46.620 align:middle line:84% There are also prepositional phrases 01:00:46.620 --> 01:00:48.540 align:middle line:84% that seem to just contain a preposition, 01:00:48.540 --> 01:00:53.460 align:middle line:84% like, "I will look up," where, again, if I say "I 01:00:53.460 --> 01:00:56.100 align:middle line:84% will look up" and you're amazed, you can say, "Up?" 01:00:56.100 --> 01:01:01.500 align:middle line:84% (Why will you look up?) "I said I would look up, and up 01:01:01.500 --> 01:01:02.130 align:middle line:90% I will look"-- 01:01:02.130 --> 01:01:04.960 align:middle line:90% 01:01:04.960 --> 01:01:07.480 align:middle line:90% Maybe. 01:01:07.480 --> 01:01:08.230 align:middle line:90% Yeah? 01:01:08.230 --> 01:01:09.647 align:middle line:84% AUDIENCE: How do we know that "up" 01:01:09.647 --> 01:01:12.983 align:middle line:84% isn't like an adverb describing "look"? 01:01:12.983 --> 01:01:13.900 align:middle line:90% NORVIN RICHARDS: Yeah. 01:01:13.900 --> 01:01:18.230 align:middle line:90% 01:01:18.230 --> 01:01:20.340 align:middle line:90% Yep. 01:01:20.340 --> 01:01:23.140 align:middle line:90% Yep, yep, yep. 01:01:23.140 --> 01:01:24.790 align:middle line:90% Yes sir? 01:01:24.790 --> 01:01:25.450 align:middle line:90% Adverb. 01:01:25.450 --> 01:01:29.350 align:middle line:84% Adverb is a funny word because there 01:01:29.350 --> 01:01:34.640 align:middle line:84% are a lot of things that can be used. 01:01:34.640 --> 01:01:39.950 align:middle line:84% So if by adverb, we mean thing that modifies the verb, 01:01:39.950 --> 01:01:44.150 align:middle line:84% there are a lot of things that can be used that way, including 01:01:44.150 --> 01:01:48.440 align:middle line:84% adverbs like "I will leave quickly," where "quickly" 01:01:48.440 --> 01:01:50.270 align:middle line:90% is an adverb. 01:01:50.270 --> 01:01:53.450 align:middle line:84% But also prepositional phrases like 01:01:53.450 --> 01:01:57.920 align:middle line:84% "I will leave in a chariot" where the prepositional phrase 01:01:57.920 --> 01:02:01.550 align:middle line:90% is like an adverb. 01:02:01.550 --> 01:02:06.800 align:middle line:84% Or maybe even noun phrases like "I 01:02:06.800 --> 01:02:09.170 align:middle line:84% will leave the day after tomorrow," 01:02:09.170 --> 01:02:13.040 align:middle line:84% where "the day after tomorrow" sure looks like a noun phrase. 01:02:13.040 --> 01:02:17.030 align:middle line:84% It's got a noun in it, "day," and then "the" before that, 01:02:17.030 --> 01:02:20.790 align:middle line:84% but we know that can go at the beginnings of noun phrases. 01:02:20.790 --> 01:02:24.350 align:middle line:84% So there are probably-- so the word "adverb" 01:02:24.350 --> 01:02:27.128 align:middle line:84% can be used to cover a bunch of things, 01:02:27.128 --> 01:02:29.420 align:middle line:84% including things that we don't have any other word for, 01:02:29.420 --> 01:02:33.080 align:middle line:84% like "quickly," which is an adverb, but also things 01:02:33.080 --> 01:02:34.790 align:middle line:84% that we would give other labels to 01:02:34.790 --> 01:02:37.410 align:middle line:84% like "noun phrase" or "prepositional phrase." 01:02:37.410 --> 01:02:39.260 align:middle line:84% And so I think you might be right that this 01:02:39.260 --> 01:02:42.680 align:middle line:84% is an adverb in the sense that it modifies the verb. 01:02:42.680 --> 01:02:44.660 align:middle line:84% I think I might also be right in the sense 01:02:44.660 --> 01:02:46.910 align:middle line:84% that it's a prepositional phrase that's 01:02:46.910 --> 01:02:48.770 align:middle line:90% being used as adverbially. 01:02:48.770 --> 01:02:49.680 align:middle line:90% Does that make sense? 01:02:49.680 --> 01:02:51.800 align:middle line:84% So what I'm trying to do is work out a way for you 01:02:51.800 --> 01:02:56.240 align:middle line:84% and I to both be right, which is always my goal in discussions. 01:02:56.240 --> 01:02:57.400 align:middle line:90% Yes? 01:02:57.400 --> 01:02:59.980 align:middle line:84% AUDIENCE: What about a sentence like "I will wake up"? 01:02:59.980 --> 01:03:01.600 align:middle line:84% NORVIN RICHARDS: "I will wake up." 01:03:01.600 --> 01:03:04.880 align:middle line:84% AUDIENCE: Does it make sense to say [INAUDIBLE]?? 01:03:04.880 --> 01:03:06.760 align:middle line:84% NORVIN RICHARDS: Yeah, that's a nice example. 01:03:06.760 --> 01:03:07.600 align:middle line:90% "I will wake up." 01:03:07.600 --> 01:03:10.630 align:middle line:84% So probably-- I mean, I just tried to convince everybody 01:03:10.630 --> 01:03:12.010 align:middle line:90% that it was OK to say up-- 01:03:12.010 --> 01:03:14.440 align:middle line:84% I said, "I would look up and up I will look." 01:03:14.440 --> 01:03:17.880 align:middle line:84% But "I said I would wake up and up I will wake," 01:03:17.880 --> 01:03:20.920 align:middle line:90% I'm not even going to try. 01:03:20.920 --> 01:03:25.060 align:middle line:84% And similarly, I think, if I say "I will wake up" 01:03:25.060 --> 01:03:28.180 align:middle line:84% and you're amazed, you're not going to say, "Up?" 01:03:28.180 --> 01:03:32.300 align:middle line:90% 01:03:32.300 --> 01:03:33.123 align:middle line:90% Yeah. 01:03:33.123 --> 01:03:34.040 align:middle line:90% That's a nice example. 01:03:34.040 --> 01:03:39.980 align:middle line:90% 01:03:39.980 --> 01:03:44.120 align:middle line:84% There are a bunch of things that look like prepositions that 01:03:44.120 --> 01:03:47.750 align:middle line:84% combine in an interesting way with verbs in English 01:03:47.750 --> 01:03:49.610 align:middle line:84% and a lot of other Germanic languages. 01:03:49.610 --> 01:03:51.290 align:middle line:90% And "wake up" is one of those. 01:03:51.290 --> 01:03:55.830 align:middle line:90% 01:03:55.830 --> 01:03:58.310 align:middle line:84% And this is actually the kind of example 01:03:58.310 --> 01:04:02.000 align:middle line:84% that I was exploiting in the first slides I was showing you 01:04:02.000 --> 01:04:07.610 align:middle line:84% about syntax, even the ones on the first day 01:04:07.610 --> 01:04:09.350 align:middle line:90% that you can say-- 01:04:09.350 --> 01:04:12.410 align:middle line:84% so wake up, you can say "I will wake up." 01:04:12.410 --> 01:04:16.790 align:middle line:90% 01:04:16.790 --> 01:04:23.885 align:middle line:84% You can also say, "I will wake up the cats." 01:04:23.885 --> 01:04:28.070 align:middle line:84% It's unwise but it's grammatical. 01:04:28.070 --> 01:04:33.140 align:middle line:84% But I think "up the cats" is not a prepositional phrase. 01:04:33.140 --> 01:04:37.198 align:middle line:84% Notice, for example, that if I say I will wake up the cats 01:04:37.198 --> 01:04:38.990 align:middle line:84% and you're astonished as you might well be, 01:04:38.990 --> 01:04:41.510 align:middle line:90% you can't say "Up the cats?" 01:04:41.510 --> 01:04:45.290 align:middle line:84% And "I said I would wake up the cats and up the cats 01:04:45.290 --> 01:04:47.660 align:middle line:90% I will wake," this is no good. 01:04:47.660 --> 01:04:48.440 align:middle line:90% Yes? 01:04:48.440 --> 01:04:51.080 align:middle line:84% AUDIENCE: I think that's not because kind of combined 01:04:51.080 --> 01:04:53.090 align:middle line:84% with the verb you can say "I will wake up" 01:04:53.090 --> 01:04:54.850 align:middle line:84% but you can't say, "I will wake down." 01:04:54.850 --> 01:04:55.360 align:middle line:90% NORVIN RICHARDS: Yeah. 01:04:55.360 --> 01:04:56.150 align:middle line:90% Yeah, yeah, yeah. 01:04:56.150 --> 01:04:56.650 align:middle line:90% Yeah. 01:04:56.650 --> 01:04:59.560 align:middle line:84% So I think-- we don't want to think of "up the cats" 01:04:59.560 --> 01:05:00.130 align:middle line:90% as a unit. 01:05:00.130 --> 01:05:02.980 align:middle line:84% We want to think of "wake up" as a unit 01:05:02.980 --> 01:05:05.408 align:middle line:90% that has the cats as an object. 01:05:05.408 --> 01:05:06.950 align:middle line:84% So this is the point you were making, 01:05:06.950 --> 01:05:09.450 align:middle line:84% and I think that you are also making, that there's something 01:05:09.450 --> 01:05:11.240 align:middle line:90% special about that interaction. 01:05:11.240 --> 01:05:11.980 align:middle line:90% Yeah. 01:05:11.980 --> 01:05:14.830 align:middle line:84% AUDIENCE: Is that an interesting pair, "Are you up for lunch?" 01:05:14.830 --> 01:05:16.150 align:middle line:90% or "Are you down for lunch?" 01:05:16.150 --> 01:05:17.424 align:middle line:90% They mean the same thing. 01:05:17.424 --> 01:05:19.722 align:middle line:90% 01:05:19.722 --> 01:05:21.430 align:middle line:84% NORVIN RICHARDS: And so we are learning-- 01:05:21.430 --> 01:05:23.770 align:middle line:84% this is why I'm glad I'm not a physicist, yes. 01:05:23.770 --> 01:05:26.710 align:middle line:84% So I work in the domain in which up and down can 01:05:26.710 --> 01:05:27.620 align:middle line:90% be the same thing. 01:05:27.620 --> 01:05:28.960 align:middle line:90% Yeah. 01:05:28.960 --> 01:05:31.720 align:middle line:84% If I were a physicist, then if NASA were to hire me, 01:05:31.720 --> 01:05:35.560 align:middle line:84% the spacecraft would have all kinds of problems. 01:05:35.560 --> 01:05:37.150 align:middle line:90% That's a nice example. 01:05:37.150 --> 01:05:40.540 align:middle line:84% Relatedly, "I will wake up the cat" is OK. 01:05:40.540 --> 01:05:46.180 align:middle line:84% You can also say, "I will wake the cats up," 01:05:46.180 --> 01:05:53.485 align:middle line:84% which is different from "I will walk up the stairs." 01:05:53.485 --> 01:05:56.560 align:middle line:90% 01:05:56.560 --> 01:06:00.460 align:middle line:84% You cannot say, "I will walk the stairs up," I think. 01:06:00.460 --> 01:06:04.700 align:middle line:84% And relatedly-- I exploited this in an earlier slide-- 01:06:04.700 --> 01:06:09.130 align:middle line:84% you can say, "I will wake him up." 01:06:09.130 --> 01:06:10.330 align:middle line:90% "I will wake them up." 01:06:10.330 --> 01:06:12.910 align:middle line:90% 01:06:12.910 --> 01:06:19.900 align:middle line:84% But "I will walk them up," no good. 01:06:19.900 --> 01:06:21.680 align:middle line:84% In fact, it's the other way around. 01:06:21.680 --> 01:06:22.390 align:middle line:90% I think, right? 01:06:22.390 --> 01:06:23.448 align:middle line:90% "I will walk them up." 01:06:23.448 --> 01:06:25.240 align:middle line:84% What are you going to do with those stairs? 01:06:25.240 --> 01:06:26.980 align:middle line:90% "I'm going to walk them up." 01:06:26.980 --> 01:06:28.300 align:middle line:90% No. 01:06:28.300 --> 01:06:30.700 align:middle line:84% You have to say I will walk up them. 01:06:30.700 --> 01:06:33.190 align:middle line:84% You cannot say, "I will wake up them." 01:06:33.190 --> 01:06:35.440 align:middle line:84% You have to say, "I will wake them up." 01:06:35.440 --> 01:06:38.170 align:middle line:90% So these "up"s are different. 01:06:38.170 --> 01:06:41.380 align:middle line:84% So "up the stairs" is a prepositional phrase. 01:06:41.380 --> 01:06:43.480 align:middle line:84% "Up the cats" is not a prepositional phrase. 01:06:43.480 --> 01:06:46.180 align:middle line:84% We need different structures for these verb phrases, 01:06:46.180 --> 01:06:48.040 align:middle line:90% and we will develop them. 01:06:48.040 --> 01:06:48.870 align:middle line:90% You had a-- 01:06:48.870 --> 01:06:49.495 align:middle line:90% AUDIENCE: Yeah. 01:06:49.495 --> 01:06:51.430 align:middle line:84% I was also going to say, but you can say-- 01:06:51.430 --> 01:06:53.222 align:middle line:84% if you're taking someone home, you can say, 01:06:53.222 --> 01:06:55.870 align:middle line:84% "I will walk you up to your room." 01:06:55.870 --> 01:06:57.340 align:middle line:90% NORVIN RICHARDS: Yes. 01:06:57.340 --> 01:06:59.110 align:middle line:84% "I will walk you up to your room." 01:06:59.110 --> 01:07:04.570 align:middle line:84% You can say, "I will walk the student up to her room." 01:07:04.570 --> 01:07:08.890 align:middle line:84% You can't say, "I will walk up the student to her room." 01:07:08.890 --> 01:07:10.393 align:middle line:90% Can you? 01:07:10.393 --> 01:07:12.310 align:middle line:84% AUDIENCE: But yeah, so it's like the opposite. 01:07:12.310 --> 01:07:13.227 align:middle line:90% NORVIN RICHARDS: Wait. 01:07:13.227 --> 01:07:15.910 align:middle line:84% Can you say, "I will walk up to the student to her room?" 01:07:15.910 --> 01:07:16.577 align:middle line:90% AUDIENCE: No. 01:07:16.577 --> 01:07:17.410 align:middle line:90% NORVIN RICHARDS: No. 01:07:17.410 --> 01:07:19.930 align:middle line:84% AUDIENCE: That implies you're walking on the student, which-- 01:07:19.930 --> 01:07:20.410 align:middle line:90% NORVIN RICHARDS: Yeah, OK. 01:07:20.410 --> 01:07:22.000 align:middle line:84% So if the student is lying down, yeah, 01:07:22.000 --> 01:07:25.060 align:middle line:84% and I'm using her as a ladder then, yes. 01:07:25.060 --> 01:07:29.620 align:middle line:84% But if not-- so I think "I will walk the student up 01:07:29.620 --> 01:07:32.440 align:middle line:84% to her room" is different from all of these, 01:07:32.440 --> 01:07:34.030 align:middle line:90% actually, kind of interestingly. 01:07:34.030 --> 01:07:38.680 align:middle line:84% I think maybe "up" is modifying "to her room." 01:07:38.680 --> 01:07:40.960 align:middle line:84% We want there to be a constituent "up to her room." 01:07:40.960 --> 01:07:44.800 align:middle line:84% Notice that if I say, "I will walk her up to her room" 01:07:44.800 --> 01:07:47.800 align:middle line:84% and you're amazed, you can say, "Up to her room?" 01:07:47.800 --> 01:07:49.960 align:middle line:84% which suggests that that's a constituent. 01:07:49.960 --> 01:07:53.440 align:middle line:90% 01:07:53.440 --> 01:07:56.800 align:middle line:84% Yeah, that's a nice third thing I should talk about when 01:07:56.800 --> 01:07:57.760 align:middle line:90% I'm talking about this. 01:07:57.760 --> 01:08:00.620 align:middle line:90% Cool. 01:08:00.620 --> 01:08:01.885 align:middle line:90% Oh, yeah? 01:08:01.885 --> 01:08:03.280 align:middle line:84% AUDIENCE: Yeah, I had a question. 01:08:03.280 --> 01:08:06.140 align:middle line:84% So I think there's something different between waking 01:08:06.140 --> 01:08:08.240 align:middle line:90% up or walking up. 01:08:08.240 --> 01:08:12.130 align:middle line:84% In a sense, you could say, [INAUDIBLE] and that 01:08:12.130 --> 01:08:13.510 align:middle line:90% would make sense. 01:08:13.510 --> 01:08:16.500 align:middle line:84% So it's kind of like, I don't know, 01:08:16.500 --> 01:08:20.010 align:middle line:84% an adjective, in some sense, is describing [? state. ?] 01:08:20.010 --> 01:08:22.450 align:middle line:84% But at the same time, it's not an adjective. 01:08:22.450 --> 01:08:24.479 align:middle line:90% It's describing a verb. 01:08:24.479 --> 01:08:28.080 align:middle line:84% So I was wondering if there's any big distinction 01:08:28.080 --> 01:08:32.170 align:middle line:84% between waking up and like walking up [INAUDIBLE].. 01:08:32.170 --> 01:08:34.550 align:middle line:84% NORVIN RICHARDS: So that's a very nice example. 01:08:34.550 --> 01:08:38.560 align:middle line:84% So in a way, waking the cats up, it's 01:08:38.560 --> 01:08:41.380 align:middle line:84% a little bit like painting the cats red. 01:08:41.380 --> 01:08:45.310 align:middle line:84% Not in very many ways, but in this way. 01:08:45.310 --> 01:08:48.069 align:middle line:84% If you paint the cats red, you paint the cats. 01:08:48.069 --> 01:08:51.640 align:middle line:84% And as a result of that, the cats are red. 01:08:51.640 --> 01:08:54.354 align:middle line:84% The cats change into cats that were not red before and now 01:08:54.354 --> 01:08:55.479 align:middle line:90% they are cats that are red. 01:08:55.479 --> 01:08:58.000 align:middle line:84% Similarly, if you wake the cats up, 01:08:58.000 --> 01:08:59.710 align:middle line:90% the cats were not up before. 01:08:59.710 --> 01:09:01.660 align:middle line:84% But when you're done, the cats are up. 01:09:01.660 --> 01:09:02.740 align:middle line:90% And possibly also red. 01:09:02.740 --> 01:09:06.790 align:middle line:84% It depends on how you did it, I guess. 01:09:06.790 --> 01:09:09.130 align:middle line:84% So I think you're absolutely right that we want "wake 01:09:09.130 --> 01:09:11.887 align:middle line:90% the cats up" to have "up." 01:09:11.887 --> 01:09:12.970 align:middle line:90% "I will wake up the cats." 01:09:12.970 --> 01:09:15.609 align:middle line:84% We want "up" to not be a preposition that's 01:09:15.609 --> 01:09:18.109 align:middle line:84% combining with the cats to be a prepositional phrase. 01:09:18.109 --> 01:09:19.160 align:middle line:90% It's something else. 01:09:19.160 --> 01:09:21.370 align:middle line:84% It's like a predicate of some kind. 01:09:21.370 --> 01:09:24.160 align:middle line:84% It's like "red" in "I will paint the cats red." 01:09:24.160 --> 01:09:28.330 align:middle line:84% And yeah, as we study these verb phrases further, 01:09:28.330 --> 01:09:31.720 align:middle line:84% we'll want to have structures that give them that character. 01:09:31.720 --> 01:09:33.670 align:middle line:90% That was a very nice example. 01:09:33.670 --> 01:09:36.100 align:middle line:84% Did you have a question a while ago? 01:09:36.100 --> 01:09:36.600 align:middle line:90% No. 01:09:36.600 --> 01:09:37.710 align:middle line:90% Joseph? 01:09:37.710 --> 01:09:41.250 align:middle line:84% AUDIENCE: I think that that does fit. 01:09:41.250 --> 01:09:42.637 align:middle line:90% You can also wake the cats. 01:09:42.637 --> 01:09:44.470 align:middle line:84% NORVIN RICHARDS: Yes, you can wake the cats. 01:09:44.470 --> 01:09:45.580 align:middle line:90% You can also paint the cats. 01:09:45.580 --> 01:09:46.600 align:middle line:90% You shouldn't, but you can. 01:09:46.600 --> 01:09:47.100 align:middle line:90% Yeah. 01:09:47.100 --> 01:09:48.220 align:middle line:90% Yeah. 01:09:48.220 --> 01:09:49.447 align:middle line:90% Did you have a question? 01:09:49.447 --> 01:09:50.072 align:middle line:90% AUDIENCE: Yeah. 01:09:50.072 --> 01:09:53.520 align:middle line:84% So with the example of walking her up to her room, 01:09:53.520 --> 01:09:58.055 align:middle line:84% I think you need to say, "I will walk her up." 01:09:58.055 --> 01:09:58.930 align:middle line:90% NORVIN RICHARDS: Yes. 01:09:58.930 --> 01:09:59.530 align:middle line:90% AUDIENCE: Right? 01:09:59.530 --> 01:10:00.405 align:middle line:90% NORVIN RICHARDS: Yep. 01:10:00.405 --> 01:10:02.890 align:middle line:84% AUDIENCE: I think because the "up" is modifying 01:10:02.890 --> 01:10:05.440 align:middle line:84% "to her room," but I don't think [? so. ?] 01:10:05.440 --> 01:10:07.450 align:middle line:84% NORVIN RICHARDS: It doesn't have to, does it? 01:10:07.450 --> 01:10:08.800 align:middle line:90% No. 01:10:08.800 --> 01:10:10.120 align:middle line:90% I will walk her up. 01:10:10.120 --> 01:10:12.370 align:middle line:90% I will walk her up to her room. 01:10:12.370 --> 01:10:15.130 align:middle line:84% I will-- but similarly, you can walk the student-- 01:10:15.130 --> 01:10:17.365 align:middle line:84% so I keep going to the student from her, 01:10:17.365 --> 01:10:19.240 align:middle line:84% because there's a difference between pronouns 01:10:19.240 --> 01:10:20.650 align:middle line:90% and non-pronouns. 01:10:20.650 --> 01:10:23.560 align:middle line:84% You can wake up the cats or you can wake the cats up. 01:10:23.560 --> 01:10:26.350 align:middle line:84% You can wake them up, and you cannot wake up them. 01:10:26.350 --> 01:10:29.680 align:middle line:84% So a pronoun has to go before this kind of "up," 01:10:29.680 --> 01:10:32.440 align:middle line:84% but the cats can go on either side. 01:10:32.440 --> 01:10:34.780 align:middle line:84% So "I will walk her up," you're absolutely right. 01:10:34.780 --> 01:10:35.980 align:middle line:90% That's the way to say that. 01:10:35.980 --> 01:10:39.400 align:middle line:84% But I think you also have to walk the student up. 01:10:39.400 --> 01:10:43.790 align:middle line:84% You can't walk up the student, unless the student is lying 01:10:43.790 --> 01:10:45.200 align:middle line:90% down and you're walking on her. 01:10:45.200 --> 01:10:48.230 align:middle line:84% But if you mean you're going to walk with the student 01:10:48.230 --> 01:10:51.620 align:middle line:84% so that the two of you are up, so we 01:10:51.620 --> 01:10:54.530 align:middle line:84% want "walk the student" up to be different from "wake 01:10:54.530 --> 01:10:58.100 align:middle line:84% the cats up," and maybe also different from "walk 01:10:58.100 --> 01:10:59.550 align:middle line:90% up the stairs." 01:10:59.550 --> 01:11:02.330 align:middle line:84% Yeah, it's a third kind of thing, which, 01:11:02.330 --> 01:11:03.952 align:middle line:84% as we work further on this, I'm-- 01:11:03.952 --> 01:11:05.660 align:middle line:84% in a way, I'm glad that we're running out 01:11:05.660 --> 01:11:07.970 align:middle line:84% of time because it means that I have a week 01:11:07.970 --> 01:11:09.650 align:middle line:90% to create slides about this. 01:11:09.650 --> 01:11:13.950 align:middle line:84% But we're going to want different structures for this. 01:11:13.950 --> 01:11:15.950 align:middle line:84% You guys are making excellent points about this. 01:11:15.950 --> 01:11:17.750 align:middle line:84% So we're developing probes into the structure 01:11:17.750 --> 01:11:19.292 align:middle line:84% of the inside of the verb phrase that 01:11:19.292 --> 01:11:23.750 align:middle line:84% mean that we're going to need at least three different kinds 01:11:23.750 --> 01:11:33.740 align:middle line:84% of structures for a sequence that looks like verb, them, up, 01:11:33.740 --> 01:11:36.110 align:middle line:90% or verb, up, them. 01:11:36.110 --> 01:11:38.630 align:middle line:90% 01:11:38.630 --> 01:11:41.630 align:middle line:84% So we've got now three kinds of examples to talk about. 01:11:41.630 --> 01:11:49.040 align:middle line:84% There's "wake them up," there's "walk her up," 01:11:49.040 --> 01:11:55.620 align:middle line:84% and there's "walk up them" as in "I walked up the stairs." 01:11:55.620 --> 01:11:58.220 align:middle line:84% And that seems to be three different things that we need 01:11:58.220 --> 01:12:00.260 align:middle line:90% three different structures for. 01:12:00.260 --> 01:12:05.630 align:middle line:84% But for "walk up them" we want "up" and "them" to combine 01:12:05.630 --> 01:12:07.040 align:middle line:90% to be a propositional phrase. 01:12:07.040 --> 01:12:09.387 align:middle line:84% But for these other two, we want something else. 01:12:09.387 --> 01:12:11.720 align:middle line:84% We're going to want to circle around and try to find out 01:12:11.720 --> 01:12:12.950 align:middle line:90% what that other thing is. 01:12:12.950 --> 01:12:13.640 align:middle line:90% Yeah. 01:12:13.640 --> 01:12:15.220 align:middle line:90% Am I getting at your point? 01:12:15.220 --> 01:12:16.220 align:middle line:90% You're absolutely right. 01:12:16.220 --> 01:12:20.810 align:middle line:84% Yeah, so we need a way of covering that. 01:12:20.810 --> 01:12:21.970 align:middle line:90% Lots of people had hands. 01:12:21.970 --> 01:12:23.660 align:middle line:90% Does anybody else-- yeah? 01:12:23.660 --> 01:12:30.550 align:middle line:84% AUDIENCE: I was going to suggest something else [INAUDIBLE] Even 01:12:30.550 --> 01:12:35.310 align:middle line:90% though you can't say, walk the-- 01:12:35.310 --> 01:12:37.645 align:middle line:84% "walk the stairs up to your room." 01:12:37.645 --> 01:12:38.770 align:middle line:90% NORVIN RICHARDS: "I walk--" 01:12:38.770 --> 01:12:41.410 align:middle line:84% AUDIENCE: You could potentially say, "Climb the ladder 01:12:41.410 --> 01:12:42.220 align:middle line:90% to your room?" 01:12:42.220 --> 01:12:43.137 align:middle line:90% NORVIN RICHARDS: Yeah. 01:12:43.137 --> 01:12:45.550 align:middle line:84% That's because you can climb a ladder. 01:12:45.550 --> 01:12:49.870 align:middle line:84% So you can climb a ladder, and you can go up to your room. 01:12:49.870 --> 01:12:52.180 align:middle line:84% That's the one where I think we want "up" possibly 01:12:52.180 --> 01:12:53.590 align:middle line:90% to modify to "your room." 01:12:53.590 --> 01:12:54.520 align:middle line:90% Or maybe not. 01:12:54.520 --> 01:12:57.190 align:middle line:84% Maybe this was your point, maybe we want "up" and "to your room" 01:12:57.190 --> 01:12:59.740 align:middle line:90% to be separate adverbs. 01:12:59.740 --> 01:13:01.600 align:middle line:84% But just like you can walk up to your room, 01:13:01.600 --> 01:13:03.910 align:middle line:84% you can climb a ladder up to your room. 01:13:03.910 --> 01:13:05.290 align:middle line:90% The ladder isn't-- 01:13:05.290 --> 01:13:05.680 align:middle line:90% AUDIENCE: --or the stairs. 01:13:05.680 --> 01:13:06.290 align:middle line:90% NORVIN RICHARDS: Yeah. 01:13:06.290 --> 01:13:07.000 align:middle line:90% Yeah. 01:13:07.000 --> 01:13:08.740 align:middle line:84% And similarly, to the extent that you 01:13:08.740 --> 01:13:11.690 align:middle line:84% can say "He's walking the stairs," 01:13:11.690 --> 01:13:13.330 align:middle line:84% which I think you can sort of say, 01:13:13.330 --> 01:13:16.870 align:middle line:84% it means he's on the stairs, walking. 01:13:16.870 --> 01:13:19.730 align:middle line:84% I think that's what you're doing there. 01:13:19.730 --> 01:13:22.420 align:middle line:84% I hate to do it, but I want to ruthlessly squelch everything 01:13:22.420 --> 01:13:23.620 align:middle line:84% else you guys want to say, because I 01:13:23.620 --> 01:13:25.630 align:middle line:84% have a triumphant slide that I want to show you. 01:13:25.630 --> 01:13:29.390 align:middle line:84% And then maybe we can unsquelch you and come back. 01:13:29.390 --> 01:13:33.640 align:middle line:84% So here's the idea, just so it's clearer for everybody. 01:13:33.640 --> 01:13:35.350 align:middle line:84% For the easy case, the one that I 01:13:35.350 --> 01:13:36.880 align:middle line:84% want to refocus your attention on, 01:13:36.880 --> 01:13:39.850 align:middle line:84% "I will find the book in the garage," what we want to do 01:13:39.850 --> 01:13:42.100 align:middle line:84% is construct a structure for this 01:13:42.100 --> 01:13:44.710 align:middle line:84% that's sensitive to all of the tests for structure 01:13:44.710 --> 01:13:46.210 align:middle line:84% that we've been developing, and it's 01:13:46.210 --> 01:13:50.140 align:middle line:84% going to involve putting things together via pairwise merge 01:13:50.140 --> 01:13:54.100 align:middle line:84% and creating labels for the things 01:13:54.100 --> 01:13:58.420 align:middle line:84% that we create via pairwise merge that are typically 01:13:58.420 --> 01:14:01.510 align:middle line:84% labels that come from one of the two things that we've merged. 01:14:01.510 --> 01:14:04.158 align:middle line:84% So when we merged "the book" or "the garage," 01:14:04.158 --> 01:14:05.950 align:middle line:84% we're going to create something we're going 01:14:05.950 --> 01:14:08.050 align:middle line:90% to give the label "noun" to. 01:14:08.050 --> 01:14:11.587 align:middle line:84% Or when we merge "in" together with "the garage," 01:14:11.587 --> 01:14:13.420 align:middle line:84% we're going to create what I've been calling 01:14:13.420 --> 01:14:15.462 align:middle line:84% a prepositional phrase, something that we'll give 01:14:15.462 --> 01:14:18.370 align:middle line:84% the label "p" to, saying this has the properties 01:14:18.370 --> 01:14:22.120 align:middle line:84% that it has-- it contains a preposition "in." 01:14:22.120 --> 01:14:25.690 align:middle line:84% Similarly, we'll combine "find" with "the book" 01:14:25.690 --> 01:14:29.110 align:middle line:84% and we'll combine "find the book" with "in the garage," 01:14:29.110 --> 01:14:32.110 align:middle line:84% and we'll end up with a structure like that. 01:14:32.110 --> 01:14:34.510 align:middle line:84% Is anybody shocked by this slide? 01:14:34.510 --> 01:14:36.670 align:middle line:90% You are, yes. 01:14:36.670 --> 01:14:37.900 align:middle line:90% What did you want to do? 01:14:37.900 --> 01:14:40.420 align:middle line:90% 01:14:40.420 --> 01:14:42.950 align:middle line:90% You're gesturing. 01:14:42.950 --> 01:14:45.780 align:middle line:84% You want to combine things in a different way? 01:14:45.780 --> 01:14:50.120 align:middle line:90% So actually, you may be right. 01:14:50.120 --> 01:14:52.370 align:middle line:90% No, here's a better way to say-- 01:14:52.370 --> 01:14:54.380 align:middle line:90% we may both be right. 01:14:54.380 --> 01:14:58.880 align:middle line:84% So this is an example where this way of merging things 01:14:58.880 --> 01:15:00.750 align:middle line:90% creates a well-formed structure. 01:15:00.750 --> 01:15:04.820 align:middle line:84% Notice that if I say, "I will find the book in the garage," 01:15:04.820 --> 01:15:05.600 align:middle line:90% it's OK. 01:15:05.600 --> 01:15:07.970 align:middle line:84% "I said I would find the book in the garage. 01:15:07.970 --> 01:15:10.280 align:middle line:84% And in the garage, I will find the book." 01:15:10.280 --> 01:15:13.130 align:middle line:90% 01:15:13.130 --> 01:15:15.257 align:middle line:84% That's a unit that topicalization 01:15:15.257 --> 01:15:16.340 align:middle line:90% gets to make reference to. 01:15:16.340 --> 01:15:17.690 align:middle line:90% That's a property of this tree. 01:15:17.690 --> 01:15:21.700 align:middle line:84% So our tests get to tease this out now. 01:15:21.700 --> 01:15:23.648 align:middle line:84% Now I'm regretting us being almost out of time 01:15:23.648 --> 01:15:25.190 align:middle line:84% because there are many things to say. 01:15:25.190 --> 01:15:28.070 align:middle line:84% What we're going to see next is that when 01:15:28.070 --> 01:15:31.200 align:middle line:84% we construct syntactic trees for strings of words, 01:15:31.200 --> 01:15:33.230 align:middle line:84% it's often the case that we get ambiguities 01:15:33.230 --> 01:15:36.320 align:middle line:84% like the "unlockable" ambiguity, places where there is more 01:15:36.320 --> 01:15:37.803 align:middle line:90% than one way to combine things. 01:15:37.803 --> 01:15:40.220 align:middle line:84% And I think you may be looking at this slide and thinking, 01:15:40.220 --> 01:15:42.290 align:middle line:84% oh, but there's another way to combine these words, 01:15:42.290 --> 01:15:43.457 align:middle line:90% and you're absolutely right. 01:15:43.457 --> 01:15:44.120 align:middle line:90% There is. 01:15:44.120 --> 01:15:45.770 align:middle line:84% And what we'll do is develop tests 01:15:45.770 --> 01:15:47.860 align:middle line:84% that allow us to see which way we've combined 01:15:47.860 --> 01:15:48.860 align:middle line:90% words in different ways. 01:15:48.860 --> 01:15:50.690 align:middle line:84% And we'll find cases like "unlockable" 01:15:50.690 --> 01:15:53.720 align:middle line:84% where, depending on in what order you combine things, 01:15:53.720 --> 01:15:55.830 align:middle line:84% you get different meanings, and our tests 01:15:55.830 --> 01:15:57.440 align:middle line:90% will combine with that. 01:15:57.440 --> 01:15:58.500 align:middle line:90% Yep? 01:15:58.500 --> 01:15:59.000 align:middle line:90% All right. 01:15:59.000 --> 01:15:59.810 align:middle line:90% Thanks, everybody. 01:15:59.810 --> 01:16:03.010 align:middle line:84% We will do this again on Tuesday. 01:16:03.010 --> 01:16:07.000 align:middle line:90%