WEBVTT 00:00:00.000 --> 00:00:00.980 align:middle line:90% 00:00:00.980 --> 00:00:01.960 align:middle line:90% [SQUEAKING] 00:00:01.960 --> 00:00:03.920 align:middle line:90% [RUSTLING] 00:00:03.920 --> 00:00:07.350 align:middle line:90% [CLICKING] 00:00:07.350 --> 00:00:13.230 align:middle line:90% 00:00:13.230 --> 00:00:16.510 align:middle line:84% NORVIN RICHARDS: So first some review. 00:00:16.510 --> 00:00:19.270 align:middle line:84% Last time, we were doing "unlockable," 00:00:19.270 --> 00:00:22.300 align:middle line:84% and I was trying to convince you that it's 00:00:22.300 --> 00:00:24.910 align:middle line:84% useful to think of words, at least some words, 00:00:24.910 --> 00:00:27.070 align:middle line:90% as consisting of multiple parts. 00:00:27.070 --> 00:00:29.590 align:middle line:84% We're calling these parts morphemes. 00:00:29.590 --> 00:00:33.850 align:middle line:84% And there's this process of combining morphemes, 00:00:33.850 --> 00:00:35.830 align:middle line:84% we called it merge, where you take 00:00:35.830 --> 00:00:38.110 align:middle line:84% two things, two morphemes-- not necessarily 00:00:38.110 --> 00:00:40.728 align:middle line:84% two morphemes, two things, and you put them together 00:00:40.728 --> 00:00:41.770 align:middle line:90% and you make a new thing. 00:00:41.770 --> 00:00:44.080 align:middle line:84% So you can take a morpheme like "un-" 00:00:44.080 --> 00:00:47.080 align:middle line:84% and a morpheme like "lock," and you can put them together 00:00:47.080 --> 00:00:50.810 align:middle line:90% to make a verb "unlock." 00:00:50.810 --> 00:00:53.320 align:middle line:90% And so I just corrected myself. 00:00:53.320 --> 00:00:55.420 align:middle line:84% Sometimes the two things you take are morphemes, 00:00:55.420 --> 00:00:57.280 align:middle line:84% but sometimes they are the results 00:00:57.280 --> 00:00:59.020 align:middle line:90% of previous instances of merge. 00:00:59.020 --> 00:01:01.360 align:middle line:84% So you can also take this thing "unlock" 00:01:01.360 --> 00:01:05.050 align:middle line:84% that you've made as a result of merging "un-" and "lock," 00:01:05.050 --> 00:01:08.080 align:middle line:84% and merge that with "-able" to get you "unlockable." 00:01:08.080 --> 00:01:09.850 align:middle line:84% That procedure that I just ran through 00:01:09.850 --> 00:01:11.650 align:middle line:84% is what gives you the tree on the left. 00:01:11.650 --> 00:01:13.720 align:middle line:84% So the trees there are meant to be 00:01:13.720 --> 00:01:18.040 align:middle line:84% representations of the order in which you did things. 00:01:18.040 --> 00:01:21.280 align:middle line:84% So the tree on the left is a representation 00:01:21.280 --> 00:01:24.340 align:middle line:90% of two merge operations. 00:01:24.340 --> 00:01:26.080 align:middle line:84% First, merging "un-" with "lock," 00:01:26.080 --> 00:01:28.510 align:middle line:84% and then merging "-able" with "unlock," 00:01:28.510 --> 00:01:34.390 align:middle line:84% giving you an adjective that means a functioning lock, 00:01:34.390 --> 00:01:36.230 align:middle line:90% able to be unlocked. 00:01:36.230 --> 00:01:38.660 align:middle line:84% This is all stuff we went through last time. 00:01:38.660 --> 00:01:40.820 align:middle line:84% And this is why the word "unlockable" is ambiguous 00:01:40.820 --> 00:01:43.278 align:middle line:84% because, well, there are two ways that you can assemble it, 00:01:43.278 --> 00:01:46.180 align:middle line:84% and the two ways give you different interpretations. 00:01:46.180 --> 00:01:47.930 align:middle line:84% Are there any questions about any of that? 00:01:47.930 --> 00:01:50.390 align:middle line:84% This is all stuff we did last time. 00:01:50.390 --> 00:01:50.890 align:middle line:90% OK. 00:01:50.890 --> 00:01:54.130 align:middle line:90% 00:01:54.130 --> 00:01:54.710 align:middle line:90% Yeah, OK. 00:01:54.710 --> 00:01:55.210 align:middle line:90% All right. 00:01:55.210 --> 00:01:57.940 align:middle line:84% So that's what we were doing with "unlockable." 00:01:57.940 --> 00:02:00.943 align:middle line:84% And then we also said, and we talked about this a little bit, 00:02:00.943 --> 00:02:03.610 align:middle line:84% there need to be some statements about what's called allomorphy. 00:02:03.610 --> 00:02:06.640 align:middle line:84% So sometimes you don't just peacefully put morphemes 00:02:06.640 --> 00:02:09.490 align:middle line:84% next to each other, they change as a result 00:02:09.490 --> 00:02:12.125 align:middle line:84% of being next to another morpheme. 00:02:12.125 --> 00:02:13.750 align:middle line:84% The result is what's called allomorphy. 00:02:13.750 --> 00:02:15.490 align:middle line:84% So a morpheme can have allomorphs, 00:02:15.490 --> 00:02:19.520 align:middle line:84% different forms that it has, depending on its environment. 00:02:19.520 --> 00:02:21.490 align:middle line:84% Sometimes our statements of allomorphy 00:02:21.490 --> 00:02:24.310 align:middle line:84% will have to be very specific to the morphemes in question. 00:02:24.310 --> 00:02:29.260 align:middle line:84% So we'll say that when you merge "go," together with past tense, 00:02:29.260 --> 00:02:30.760 align:middle line:90% that the result is "went." 00:02:30.760 --> 00:02:33.310 align:middle line:84% And that's not the result of a general process, that's 00:02:33.310 --> 00:02:35.500 align:middle line:84% something you just have to look up when you look up 00:02:35.500 --> 00:02:38.620 align:middle line:90% the word "go" in English. 00:02:38.620 --> 00:02:42.010 align:middle line:84% Or that when you add "-ity" to something ending in "-ic," 00:02:42.010 --> 00:02:45.400 align:middle line:84% like "electric," that the "k" sound at the end of "-ic" 00:02:45.400 --> 00:02:47.150 align:middle line:90% will be pronounced like an "s." 00:02:47.150 --> 00:02:50.800 align:middle line:84% You get "electricity" and not "electric-ity." 00:02:50.800 --> 00:02:52.600 align:middle line:84% So sometimes these are statements 00:02:52.600 --> 00:02:55.840 align:middle line:84% about particular morphemes and what they do when they combine. 00:02:55.840 --> 00:02:58.750 align:middle line:84% Other times, we'll be able to do allomorphy 00:02:58.750 --> 00:03:00.110 align:middle line:90% via more general rules. 00:03:00.110 --> 00:03:02.980 align:middle line:84% So we did spend some time talking about Polish 00:03:02.980 --> 00:03:04.900 align:middle line:84% where we convinced ourselves that it's 00:03:04.900 --> 00:03:07.820 align:middle line:84% useful to think that Polish has a general rule-- 00:03:07.820 --> 00:03:11.740 align:middle line:84% which we'll circle back and talk about more later on-- 00:03:11.740 --> 00:03:13.360 align:middle line:84% a general rule that says that if you 00:03:13.360 --> 00:03:16.270 align:middle line:84% have a "g" at the end of a word, it becomes a "k." 00:03:16.270 --> 00:03:20.800 align:middle line:84% So in Polish, there are various words 00:03:20.800 --> 00:03:23.410 align:middle line:84% that when you pronounce them, they seem to end in "k," 00:03:23.410 --> 00:03:25.960 align:middle line:84% and then we saw that if you add the suffix so that the "k" is 00:03:25.960 --> 00:03:27.970 align:middle line:84% no longer at the end of a word, sometimes 00:03:27.970 --> 00:03:30.340 align:middle line:84% it really was a "k," in which case it stays a "k." 00:03:30.340 --> 00:03:33.460 align:middle line:84% But sometimes it was a "g" concealing itself as a "k," 00:03:33.460 --> 00:03:37.060 align:middle line:84% and it reveals its true self as a "g" when you add the suffix. 00:03:37.060 --> 00:03:39.820 align:middle line:84% Yep, this is a crash course in Polish. 00:03:39.820 --> 00:03:41.690 align:middle line:84% Now you know as much about Polish as I do, 00:03:41.690 --> 00:03:43.750 align:middle line:90% which is not much. 00:03:43.750 --> 00:03:45.340 align:middle line:90% Any-- this is all review. 00:03:45.340 --> 00:03:46.450 align:middle line:90% Does this all make sense? 00:03:46.450 --> 00:03:48.490 align:middle line:90% No questions? 00:03:48.490 --> 00:03:49.030 align:middle line:90% All right. 00:03:49.030 --> 00:03:53.750 align:middle line:84% So one of the things that we did several times-- 00:03:53.750 --> 00:03:55.690 align:middle line:84% in fact, I casually did it on the last slide-- 00:03:55.690 --> 00:03:58.360 align:middle line:84% was to use words like "noun" and "adjective." 00:03:58.360 --> 00:04:01.210 align:middle line:84% And I thought we should just take a second 00:04:01.210 --> 00:04:05.500 align:middle line:84% to talk about what we mean when we say words like that. 00:04:05.500 --> 00:04:07.840 align:middle line:84% I bet that a lot of you were taught 00:04:07.840 --> 00:04:09.430 align:middle line:90% what nouns and adjectives are. 00:04:09.430 --> 00:04:11.360 align:middle line:84% But it's possible that some of you were not, 00:04:11.360 --> 00:04:14.020 align:middle line:84% and I know that some of you are not native speakers of English, 00:04:14.020 --> 00:04:17.860 align:middle line:84% and this might not be high on my list of words to learn, 00:04:17.860 --> 00:04:20.649 align:middle line:84% if I were learning another language. 00:04:20.649 --> 00:04:23.230 align:middle line:84% Wow, my power cord just turned into a pretzel. 00:04:23.230 --> 00:04:24.440 align:middle line:90% This is pretty impressive. 00:04:24.440 --> 00:04:27.200 align:middle line:90% I didn't know they did that. 00:04:27.200 --> 00:04:30.700 align:middle line:84% So I wanted to just talk briefly about that before we 00:04:30.700 --> 00:04:33.370 align:middle line:90% go any further into morphology. 00:04:33.370 --> 00:04:35.290 align:middle line:90% What's a noun? 00:04:35.290 --> 00:04:38.860 align:middle line:84% You guys, maybe some of you were taught what a noun is. 00:04:38.860 --> 00:04:39.580 align:middle line:90% Yes. 00:04:39.580 --> 00:04:41.170 align:middle line:90% An object or a thing? 00:04:41.170 --> 00:04:42.520 align:middle line:90% An object or a thing? 00:04:42.520 --> 00:04:43.390 align:middle line:90% Yeah. 00:04:43.390 --> 00:04:47.020 align:middle line:84% Anybody else taught things like that about what nouns are? 00:04:47.020 --> 00:04:47.950 align:middle line:90% Yeah. 00:04:47.950 --> 00:04:49.290 align:middle line:90% Yes. 00:04:49.290 --> 00:04:51.250 align:middle line:84% A signifier of a particular matter. 00:04:51.250 --> 00:04:53.405 align:middle line:84% A signifier of a particular matter. 00:04:53.405 --> 00:04:55.450 align:middle line:90% Huh, I like that. 00:04:55.450 --> 00:04:57.190 align:middle line:90% That's a very classy definition. 00:04:57.190 --> 00:04:57.910 align:middle line:90% Yes? 00:04:57.910 --> 00:04:59.663 align:middle line:90% AUDIENCE: [INAUDIBLE] 00:04:59.663 --> 00:05:00.580 align:middle line:90% NORVIN RICHARDS: Yeah. 00:05:00.580 --> 00:05:02.410 align:middle line:84% So I was taught that, a person-- actually, I 00:05:02.410 --> 00:05:04.285 align:middle line:84% was taught a person, place, or a thing, yeah. 00:05:04.285 --> 00:05:07.780 align:middle line:84% And then later, I learned that kids today were 00:05:07.780 --> 00:05:09.430 align:middle line:90% learning that it was also idea. 00:05:09.430 --> 00:05:12.990 align:middle line:84% I guess on the idea that ideas are not things, which, I guess, 00:05:12.990 --> 00:05:13.490 align:middle line:90% they're not. 00:05:13.490 --> 00:05:15.940 align:middle line:84% So these are all definitions of "noun" 00:05:15.940 --> 00:05:18.280 align:middle line:84% that make reference to the kinds of things 00:05:18.280 --> 00:05:23.590 align:middle line:84% that nouns mean, right, so that they refer to people, or ideas, 00:05:23.590 --> 00:05:26.620 align:middle line:84% or things, or-- yeah, this is what nouns are. 00:05:26.620 --> 00:05:30.280 align:middle line:84% And that's fine if that works for you. 00:05:30.280 --> 00:05:33.310 align:middle line:84% That might be the best way for you to think about nouns. 00:05:33.310 --> 00:05:35.240 align:middle line:84% Here's another way of thinking about it, 00:05:35.240 --> 00:05:38.800 align:middle line:84% which you can use if you want to if you're ever confused about 00:05:38.800 --> 00:05:40.600 align:middle line:84% whether something is a noun or not, 00:05:40.600 --> 00:05:43.840 align:middle line:84% if you find yourself as you go further in linguistics, 00:05:43.840 --> 00:05:45.250 align:middle line:90% every so often, you will-- 00:05:45.250 --> 00:05:46.441 align:middle line:90% yeah. 00:05:46.441 --> 00:05:48.790 align:middle line:84% Part of the point of higher education 00:05:48.790 --> 00:05:51.730 align:middle line:84% is to make you, like, doubt yourself and be unhappy, 00:05:51.730 --> 00:05:53.902 align:middle line:84% things that you were sure that you understood, 00:05:53.902 --> 00:05:55.360 align:middle line:84% all of a sudden, you're like, wait, 00:05:55.360 --> 00:05:57.370 align:middle line:84% do I really deeply understand that? 00:05:57.370 --> 00:06:01.480 align:middle line:84% And it's meant to undermine your emotional security, basically, 00:06:01.480 --> 00:06:04.030 align:middle line:90% that's what education is for. 00:06:04.030 --> 00:06:08.320 align:middle line:84% So here's another way of thinking about what nouns are. 00:06:08.320 --> 00:06:11.360 align:middle line:84% If you're making a sentence, let's say, 00:06:11.360 --> 00:06:14.120 align:middle line:84% putting a sentence together is like putting together 00:06:14.120 --> 00:06:17.320 align:middle line:84% any other complicated thing with lots of parts. 00:06:17.320 --> 00:06:19.930 align:middle line:84% Take your favorite example of that, a jigsaw puzzle, 00:06:19.930 --> 00:06:22.480 align:middle line:84% or a model airplane, or whatever, 00:06:22.480 --> 00:06:27.130 align:middle line:84% or IKEA furniture, if you've ever assembled that. 00:06:27.130 --> 00:06:30.100 align:middle line:84% If you're ever assembling things like that, 00:06:30.100 --> 00:06:33.520 align:middle line:84% they have different parts with different shapes 00:06:33.520 --> 00:06:35.140 align:middle line:84% that go in different places, right? 00:06:35.140 --> 00:06:37.720 align:middle line:84% If you're making a model airplane, there are the wheels, 00:06:37.720 --> 00:06:40.540 align:middle line:84% and the wheels go, well, where the wheels go. 00:06:40.540 --> 00:06:43.150 align:middle line:84% Or if you're making an IKEA chair, 00:06:43.150 --> 00:06:47.350 align:middle line:84% there are various boards, but then there 00:06:47.350 --> 00:06:48.850 align:middle line:90% are also screws and-- right? 00:06:48.850 --> 00:06:50.392 align:middle line:84% There are things that you're supposed 00:06:50.392 --> 00:06:52.180 align:middle line:90% to put in particular places. 00:06:52.180 --> 00:06:55.680 align:middle line:84% And you could name those things by their shapes, right? 00:06:55.680 --> 00:06:59.160 align:middle line:84% So you could talk about the wheel on an airplane, 00:06:59.160 --> 00:07:03.210 align:middle line:84% or the edge piece on a jigsaw puzzle. 00:07:03.210 --> 00:07:07.170 align:middle line:84% And you can also kind of name them by their functions. 00:07:07.170 --> 00:07:09.720 align:middle line:84% If you weren't sure what an edge piece was, 00:07:09.720 --> 00:07:11.910 align:middle line:84% well, it would be the piece that goes 00:07:11.910 --> 00:07:14.370 align:middle line:84% on the edge of the jigsaw puzzle, the one that 00:07:14.370 --> 00:07:16.110 align:middle line:90% goes on the outside. 00:07:16.110 --> 00:07:20.120 align:middle line:84% And I guess kind of similarly, "noun"-- 00:07:20.120 --> 00:07:22.640 align:middle line:84% when we say that something is a noun, what we mean 00:07:22.640 --> 00:07:25.370 align:middle line:84% is that it's something that goes where nouns 00:07:25.370 --> 00:07:28.100 align:middle line:90% go, kind of like an edge piece. 00:07:28.100 --> 00:07:30.360 align:middle line:84% It's a piece that goes on the edge. 00:07:30.360 --> 00:07:32.330 align:middle line:84% So if you're putting a sentence together, 00:07:32.330 --> 00:07:36.050 align:middle line:84% there are parts of the sentence where nouns belong. 00:07:36.050 --> 00:07:38.750 align:middle line:84% And noun is just a name for the class of things 00:07:38.750 --> 00:07:41.160 align:middle line:90% that can go in those places. 00:07:41.160 --> 00:07:45.110 align:middle line:84% So, I mean, for example, if you can grammatically 00:07:45.110 --> 00:07:47.840 align:middle line:84% finish a sentence with a single word, 00:07:47.840 --> 00:07:51.530 align:middle line:84% a sentence like, we are talking about (blank), 00:07:51.530 --> 00:07:58.210 align:middle line:84% or we are talking about "the" (blank), then it's a noun. 00:07:58.210 --> 00:08:01.755 align:middle line:84% And I'm saying this partly as a challenge, try to show me 00:08:01.755 --> 00:08:02.380 align:middle line:90% that I'm wrong. 00:08:02.380 --> 00:08:02.880 align:middle line:90% Yes. 00:08:02.880 --> 00:08:04.900 align:middle line:90% AUDIENCE: [INAUDIBLE]. 00:08:04.900 --> 00:08:07.210 align:middle line:84% NORVIN RICHARDS: You don't, you're right. 00:08:07.210 --> 00:08:09.440 align:middle line:90% So we are talking about "him." 00:08:09.440 --> 00:08:11.500 align:middle line:84% So "him" gets to count as a noun. 00:08:11.500 --> 00:08:15.310 align:middle line:84% And we would want pronouns to be a particular subspecies 00:08:15.310 --> 00:08:17.830 align:middle line:90% of noun. 00:08:17.830 --> 00:08:20.260 align:middle line:84% That might not be such a bad result. Yes. 00:08:20.260 --> 00:08:23.693 align:middle line:84% AUDIENCE: Wouldn't names also be able to fit in that? 00:08:23.693 --> 00:08:24.610 align:middle line:90% NORVIN RICHARDS: Yeah. 00:08:24.610 --> 00:08:26.920 align:middle line:90% So we're talking about Mary. 00:08:26.920 --> 00:08:28.420 align:middle line:90% "Mary" would be a noun. 00:08:28.420 --> 00:08:29.800 align:middle line:84% And that's probably right, "Mary" 00:08:29.800 --> 00:08:31.130 align:middle line:90% should probably be a noun. 00:08:31.130 --> 00:08:31.630 align:middle line:90% Yes. 00:08:31.630 --> 00:08:34.330 align:middle line:84% AUDIENCE: In a sentence we are talking about running, 00:08:34.330 --> 00:08:36.026 align:middle line:84% there's a word for that, isn't there? 00:08:36.026 --> 00:08:37.803 align:middle line:84% NORVIN RICHARDS: It's a gerund, yeah. 00:08:37.803 --> 00:08:39.220 align:middle line:84% So this is one of the places where 00:08:39.220 --> 00:08:41.303 align:middle line:84% you're supposed to be filled with emotional angst. 00:08:41.303 --> 00:08:42.909 align:middle line:84% Wait, "running," it's got "run" in it. 00:08:42.909 --> 00:08:44.030 align:middle line:90% That looks like a verb. 00:08:44.030 --> 00:08:46.640 align:middle line:84% But it's a verb that's been converted into a noun 00:08:46.640 --> 00:08:48.460 align:middle line:84% so that we can put it in this sentence. 00:08:48.460 --> 00:08:49.960 align:middle line:90% That's what a gerund is. 00:08:49.960 --> 00:08:53.530 align:middle line:84% It's a name for a verb that's been turned into a noun. 00:08:53.530 --> 00:08:56.270 align:middle line:90% Yeah, good point. 00:08:56.270 --> 00:08:58.540 align:middle line:90% Yeah. 00:08:58.540 --> 00:08:59.040 align:middle line:90% OK. 00:08:59.040 --> 00:09:00.480 align:middle line:84% So I invite you to try to come up 00:09:00.480 --> 00:09:03.670 align:middle line:84% with sentences like this for your favorite parts of speech. 00:09:03.670 --> 00:09:04.980 align:middle line:90% Here's another one. 00:09:04.980 --> 00:09:07.710 align:middle line:84% If you can grammatically finish a sentence like, "I consider 00:09:07.710 --> 00:09:11.250 align:middle line:84% her (blank)," and the result is a sentence, that means, 00:09:11.250 --> 00:09:13.710 align:middle line:90% "I think that she is (blank)." 00:09:13.710 --> 00:09:18.550 align:middle line:84% So I consider her tall, or brilliant, or whatever, 00:09:18.550 --> 00:09:19.855 align:middle line:90% then that's an adjective. 00:09:19.855 --> 00:09:21.105 align:middle line:90% The thing that's in the blank. 00:09:21.105 --> 00:09:24.120 align:middle line:90% 00:09:24.120 --> 00:09:27.030 align:middle line:84% Sometimes the sentences will not make a whole lot of sense. 00:09:27.030 --> 00:09:30.390 align:middle line:90% So "I consider her autumnal"-- 00:09:30.390 --> 00:09:32.400 align:middle line:90% It's not clear what that means. 00:09:32.400 --> 00:09:37.380 align:middle line:84% But if you compare it with "I consider her run," yeah. 00:09:37.380 --> 00:09:39.880 align:middle line:84% So "I consider her run" just makes no sense at all. 00:09:39.880 --> 00:09:41.910 align:middle line:84% Whereas "I consider her autumnal." 00:09:41.910 --> 00:09:44.170 align:middle line:90% I sound deeply poetic, I guess. 00:09:44.170 --> 00:09:45.790 align:middle line:84% And you wonder what I mean by that. 00:09:45.790 --> 00:09:46.290 align:middle line:90% Yeah. 00:09:46.290 --> 00:09:48.060 align:middle line:84% But-- so "autumnal" is an adjective, 00:09:48.060 --> 00:09:50.010 align:middle line:90% "run" is not an adjective. 00:09:50.010 --> 00:09:52.980 align:middle line:84% And I have this bit about that it has to mean, 00:09:52.980 --> 00:09:56.310 align:middle line:84% "I think she is (blank)," because you can say things 00:09:56.310 --> 00:09:58.560 align:middle line:90% like, "I consider her often." 00:09:58.560 --> 00:10:00.630 align:middle line:84% That means "I think about her often." 00:10:00.630 --> 00:10:02.760 align:middle line:84% "Often" is not an adjective, because if I 00:10:02.760 --> 00:10:04.590 align:middle line:84% say I consider her often, I don't 00:10:04.590 --> 00:10:06.450 align:middle line:90% mean I think that she is often. 00:10:06.450 --> 00:10:07.020 align:middle line:90% Yeah. 00:10:07.020 --> 00:10:09.300 align:middle line:90% That's not what that means. 00:10:09.300 --> 00:10:10.750 align:middle line:90% So "often" is not an adjective. 00:10:10.750 --> 00:10:13.230 align:middle line:84% It's an adverb, actually, and we'll talk about that. 00:10:13.230 --> 00:10:14.958 align:middle line:84% I'm not going to make you go through this 00:10:14.958 --> 00:10:16.500 align:middle line:84% for all the parts of speech, but this 00:10:16.500 --> 00:10:17.520 align:middle line:90% is an interesting exercise. 00:10:17.520 --> 00:10:18.030 align:middle line:90% Yes. 00:10:18.030 --> 00:10:19.875 align:middle line:84% AUDIENCE: Is this the answer to our problem? 00:10:19.875 --> 00:10:21.450 align:middle line:90% NORVIN RICHARDS: Uh-huh. 00:10:21.450 --> 00:10:22.260 align:middle line:90% AUDIENCE: Yes. 00:10:22.260 --> 00:10:26.125 align:middle line:84% It's also [INAUDIBLE],, I think it's coming [INAUDIBLE].. 00:10:26.125 --> 00:10:27.750 align:middle line:84% NORVIN RICHARDS: I consider her coming. 00:10:27.750 --> 00:10:30.600 align:middle line:90% 00:10:30.600 --> 00:10:34.380 align:middle line:90% I'm full of doubt and angst. 00:10:34.380 --> 00:10:38.040 align:middle line:84% So there is a parse of that, I consider her coming, which 00:10:38.040 --> 00:10:41.820 align:middle line:84% means, I am thinking about the fact that she is coming. 00:10:41.820 --> 00:10:44.490 align:middle line:90% It can mean that, I think. 00:10:44.490 --> 00:10:48.825 align:middle line:84% I consider her-- can it mean I consid-- 00:10:48.825 --> 00:10:50.265 align:middle line:90% I believe that she is coming? 00:10:50.265 --> 00:10:52.832 align:middle line:90% 00:10:52.832 --> 00:10:55.040 align:middle line:84% I know there are other English speakers in this room. 00:10:55.040 --> 00:10:55.940 align:middle line:90% What do you guys think? 00:10:55.940 --> 00:10:56.440 align:middle line:90% Yes. 00:10:56.440 --> 00:10:59.660 align:middle line:84% AUDIENCE: I know there's like that one part of the game not 00:10:59.660 --> 00:11:10.833 align:middle line:90% really ever [INAUDIBLE] 00:11:10.833 --> 00:11:11.750 align:middle line:90% NORVIN RICHARDS: Yeah. 00:11:11.750 --> 00:11:13.151 align:middle line:90% AUDIENCE: [INAUDIBLE]. 00:11:13.151 --> 00:11:16.887 align:middle line:90% 00:11:16.887 --> 00:11:20.830 align:middle line:90% NORVIN RICHARDS: Oh, yeah. 00:11:20.830 --> 00:11:21.330 align:middle line:90% Yeah. 00:11:21.330 --> 00:11:25.083 align:middle line:90% So-- but I think-- 00:11:25.083 --> 00:11:26.250 align:middle line:90% I'm sorry, what's your name? 00:11:26.250 --> 00:11:27.170 align:middle line:90% Is it Ivana? 00:11:27.170 --> 00:11:28.050 align:middle line:90% No. 00:11:28.050 --> 00:11:29.700 align:middle line:84% Sorry, what is your-- what is your-- 00:11:29.700 --> 00:11:32.010 align:middle line:84% you were asking about "I consider her coming" 00:11:32.010 --> 00:11:33.155 align:middle line:90% just a second ago. 00:11:33.155 --> 00:11:33.655 align:middle line:90% Yeah. 00:11:33.655 --> 00:11:34.488 align:middle line:90% AUDIENCE: Oh, Vlada. 00:11:34.488 --> 00:11:37.470 align:middle line:84% NORVIN RICHARDS: Sorry, Vlada, I got you wrong, sorry. 00:11:37.470 --> 00:11:39.270 align:middle line:84% Vlada is asking about whether it's 00:11:39.270 --> 00:11:43.710 align:middle line:84% possible to have a form like "coming" as an adjective here. 00:11:43.710 --> 00:11:45.940 align:middle line:84% Forms like "coming" can be adjectives. 00:11:45.940 --> 00:11:48.480 align:middle line:84% So you can say things like "the coming century," 00:11:48.480 --> 00:11:51.970 align:middle line:84% right, where "coming" is being used as an adjective. 00:11:51.970 --> 00:11:56.910 align:middle line:84% So I said a second ago, "running" can be a noun, right? 00:11:56.910 --> 00:11:58.830 align:middle line:90% We're talking about "running." 00:11:58.830 --> 00:12:02.280 align:middle line:84% This is one of many tricky things about English. 00:12:02.280 --> 00:12:05.670 align:middle line:84% Forms of verbs that end in "-ing" can be nouns, 00:12:05.670 --> 00:12:07.080 align:middle line:90% and then we call them gerunds. 00:12:07.080 --> 00:12:10.170 align:middle line:84% They can also be participles, and then we 00:12:10.170 --> 00:12:12.180 align:middle line:90% call them adjectives. 00:12:12.180 --> 00:12:13.883 align:middle line:84% Sorry, they can also be adjectives, 00:12:13.883 --> 00:12:15.300 align:middle line:84% and then we call them participles. 00:12:15.300 --> 00:12:16.367 align:middle line:90% Let me do that again. 00:12:16.367 --> 00:12:18.450 align:middle line:84% "-ing," if you add it to a noun, there's an "-ing" 00:12:18.450 --> 00:12:21.150 align:middle line:84% that you add to a noun that makes it into a-- sorry. 00:12:21.150 --> 00:12:23.400 align:middle line:84% There's an "-ing" that you add to a verb that makes it 00:12:23.400 --> 00:12:24.690 align:middle line:90% into a noun. 00:12:24.690 --> 00:12:27.150 align:middle line:84% That's when it's a gerund, like in "We're 00:12:27.150 --> 00:12:28.110 align:middle line:90% talking about running." 00:12:28.110 --> 00:12:30.360 align:middle line:84% There's an "-ing" that you add to a verb that makes it 00:12:30.360 --> 00:12:32.580 align:middle line:84% into an adjective, and that's a present participle. 00:12:32.580 --> 00:12:34.538 align:middle line:84% That's the kind of thing Vlada is asking about. 00:12:34.538 --> 00:12:37.260 align:middle line:84% So "the coming century," or "a running man," 00:12:37.260 --> 00:12:41.010 align:middle line:84% those 'running"s are-- those "-ing" verbs are adjectives, 00:12:41.010 --> 00:12:43.080 align:middle line:90% they're participles. 00:12:43.080 --> 00:12:45.450 align:middle line:84% Sort of like, if we go back to "un-" 00:12:45.450 --> 00:12:47.265 align:middle line:84% we convinced ourselves that "un-" 00:12:47.265 --> 00:12:51.840 align:middle line:84% has a couple of different forms, that it can attach either 00:12:51.840 --> 00:12:54.240 align:middle line:84% to an adjective, or to a verb, right, 00:12:54.240 --> 00:12:55.620 align:middle line:84% with slightly different meanings. 00:12:55.620 --> 00:13:00.360 align:middle line:84% Similarly, "-ing" has several different guises. 00:13:00.360 --> 00:13:02.160 align:middle line:84% It can be a couple of different things. 00:13:02.160 --> 00:13:04.050 align:middle line:84% English really needs more morphemes. 00:13:04.050 --> 00:13:06.982 align:middle line:84% We have several morphemes that are pronounced the same way. 00:13:06.982 --> 00:13:08.190 align:middle line:90% Not every language does that. 00:13:08.190 --> 00:13:09.607 align:middle line:84% There are plenty of languages that 00:13:09.607 --> 00:13:13.890 align:middle line:84% have just different forms for gerunds and participles. 00:13:13.890 --> 00:13:14.410 align:middle line:90% OK. 00:13:14.410 --> 00:13:14.910 align:middle line:90% All right. 00:13:14.910 --> 00:13:16.440 align:middle line:84% So this has just been an exercise 00:13:16.440 --> 00:13:18.780 align:middle line:84% in trying to convince ourselves that we know what we're talking 00:13:18.780 --> 00:13:19.860 align:middle line:84% about when we're talking about things 00:13:19.860 --> 00:13:21.270 align:middle line:90% like nouns and adjectives. 00:13:21.270 --> 00:13:24.060 align:middle line:84% And like I say, I won't make you do 00:13:24.060 --> 00:13:27.100 align:middle line:90% this for every part of speech. 00:13:27.100 --> 00:13:30.990 align:middle line:84% But if you want to annoy your friends back 00:13:30.990 --> 00:13:35.070 align:middle line:84% in the dorm, trying to come up with sentences like this 00:13:35.070 --> 00:13:36.880 align:middle line:84% might be an interesting exercise. 00:13:36.880 --> 00:13:38.050 align:middle line:90% Yeah. 00:13:38.050 --> 00:13:41.270 align:middle line:84% Interesting for me, maybe not for them. 00:13:41.270 --> 00:13:41.770 align:middle line:90% OK. 00:13:41.770 --> 00:13:43.880 align:middle line:90% Questions about any of that? 00:13:43.880 --> 00:13:45.830 align:middle line:84% This has just been so that if anybody 00:13:45.830 --> 00:13:47.930 align:middle line:84% heard all this talk about nouns and adjectives 00:13:47.930 --> 00:13:50.815 align:middle line:84% and was wondering what the heck are we talking about exactly, 00:13:50.815 --> 00:13:52.565 align:middle line:84% that is what we are talking about exactly. 00:13:52.565 --> 00:13:55.760 align:middle line:90% 00:13:55.760 --> 00:13:59.570 align:middle line:84% So we've been talking now about morphology for a little while. 00:13:59.570 --> 00:14:03.740 align:middle line:84% And we have mostly been talking about language 00:14:03.740 --> 00:14:05.310 align:middle line:90% specific properties. 00:14:05.310 --> 00:14:09.620 align:middle line:84% So we've been saying a morpheme with a given meaning 00:14:09.620 --> 00:14:11.730 align:middle line:84% is pronounced differently in different languages, 00:14:11.730 --> 00:14:12.980 align:middle line:90% so we started with cats. 00:14:12.980 --> 00:14:15.410 align:middle line:84% We convinced ourselves that if we 00:14:15.410 --> 00:14:20.870 align:middle line:84% wanted to look for things that were universal about language, 00:14:20.870 --> 00:14:22.520 align:middle line:84% figuring out that the word for cat 00:14:22.520 --> 00:14:25.280 align:middle line:84% is "cat," that's clearly not universal. 00:14:25.280 --> 00:14:28.280 align:middle line:84% Every language has a different word for cat. 00:14:28.280 --> 00:14:30.380 align:middle line:84% So morphemes with a different meaning 00:14:30.380 --> 00:14:34.760 align:middle line:84% can be pronounced differently in different languages. 00:14:34.760 --> 00:14:38.220 align:middle line:90% Morphemes can be various types. 00:14:38.220 --> 00:14:39.980 align:middle line:84% They can be prefixes, or suffixes, 00:14:39.980 --> 00:14:41.730 align:middle line:90% or infixes, or other things. 00:14:41.730 --> 00:14:46.580 align:middle line:84% So those here are three verbs in English, Lardil, and Tagalog, 00:14:46.580 --> 00:14:48.800 align:middle line:90% that all mean "danced." 00:14:48.800 --> 00:14:51.470 align:middle line:84% They all have-- consist of a verb, 00:14:51.470 --> 00:14:54.050 align:middle line:84% along with another morpheme that indicates past tense. 00:14:54.050 --> 00:14:55.700 align:middle line:84% That morpheme is a suffix in English, 00:14:55.700 --> 00:14:59.930 align:middle line:84% but it's a prefix in Lardil, and it's an infix in Tagalog. 00:14:59.930 --> 00:15:03.320 align:middle line:84% And morphemes can be bound or free. 00:15:03.320 --> 00:15:06.390 align:middle line:84% So here are some new examples of that. 00:15:06.390 --> 00:15:08.640 align:middle line:84% This is, again, a point of cross-linguistic variation. 00:15:08.640 --> 00:15:12.450 align:middle line:84% So in English, a phrase like "in my hand" 00:15:12.450 --> 00:15:16.130 align:middle line:84% consists of three words, three free morphemes. 00:15:16.130 --> 00:15:21.540 align:middle line:84% But in Turkish, it consists of a root plus two bound morphemes. 00:15:21.540 --> 00:15:24.890 align:middle line:84% So there's the word for hand, to which you 00:15:24.890 --> 00:15:28.050 align:middle line:84% are adding suffixes that mean things like "my" and "in." 00:15:28.050 --> 00:15:28.550 align:middle line:90% Yeah. 00:15:28.550 --> 00:15:30.150 align:middle line:90% Anybody here speak Turkish? 00:15:30.150 --> 00:15:30.650 align:middle line:90% Cool. 00:15:30.650 --> 00:15:31.400 align:middle line:90% Is this true? 00:15:31.400 --> 00:15:32.120 align:middle line:90% Am I lying? 00:15:32.120 --> 00:15:35.040 align:middle line:90% OK, good. 00:15:35.040 --> 00:15:36.950 align:middle line:84% Similarly, in English, a sentence 00:15:36.950 --> 00:15:40.520 align:middle line:84% like "I bought a bed" consists of four free standing words. 00:15:40.520 --> 00:15:43.200 align:middle line:84% In Mohawk, it consists of one word. 00:15:43.200 --> 00:15:46.470 align:middle line:84% So there's a verb "buy" that is combining with various things, 00:15:46.470 --> 00:15:48.800 align:middle line:84% including a morpheme that means "bed." 00:15:48.800 --> 00:15:52.530 align:middle line:84% So the object becomes part of this long verb. 00:15:52.530 --> 00:15:54.600 align:middle line:90% Anybody here speak Mohawk? 00:15:54.600 --> 00:15:55.100 align:middle line:90% OK, good. 00:15:55.100 --> 00:15:56.450 align:middle line:90% So please just believe me, yeah. 00:15:56.450 --> 00:15:58.310 align:middle line:90% This is how Mohawk works. 00:15:58.310 --> 00:16:00.830 align:middle line:84% This is a process called noun incorporation. 00:16:00.830 --> 00:16:03.020 align:middle line:84% It's not obligatory in Mohawk, but it's an option 00:16:03.020 --> 00:16:05.360 align:middle line:90% that you have. 00:16:05.360 --> 00:16:06.830 align:middle line:84% So languages can vary with respect 00:16:06.830 --> 00:16:10.100 align:middle line:84% to how their morphemes are pronounced, 00:16:10.100 --> 00:16:12.780 align:middle line:84% whether they're prefixes or suffixes, or other things, 00:16:12.780 --> 00:16:16.310 align:middle line:84% and whether they are bound or free. 00:16:16.310 --> 00:16:19.700 align:middle line:84% Languages are sometimes informally classified 00:16:19.700 --> 00:16:23.450 align:middle line:84% by how likely their morphemes are to be bound. 00:16:23.450 --> 00:16:27.650 align:middle line:84% And when I say that they're informally 00:16:27.650 --> 00:16:29.630 align:middle line:84% classified in this way, what I mean is, 00:16:29.630 --> 00:16:34.907 align:middle line:84% it's probably hopeless to try to tightly define what this means. 00:16:34.907 --> 00:16:36.740 align:middle line:84% But this is a quick and dirty classification 00:16:36.740 --> 00:16:38.030 align:middle line:90% that you will hear people use. 00:16:38.030 --> 00:16:39.650 align:middle line:90% There are languages out there-- 00:16:39.650 --> 00:16:43.820 align:middle line:84% Mandarin is the poster child for this-- that just are not 00:16:43.820 --> 00:16:45.390 align:middle line:90% into bound morphemes. 00:16:45.390 --> 00:16:48.360 align:middle line:84% So all of the morphemes are free. 00:16:48.360 --> 00:16:50.480 align:middle line:84% They don't have to be attached to anything. 00:16:50.480 --> 00:16:53.960 align:middle line:84% So in Chinese, "he ate the meal" consists 00:16:53.960 --> 00:16:59.450 align:middle line:84% of four free standing words, doesn't have prefixes 00:16:59.450 --> 00:17:01.280 align:middle line:90% or suffixes, mostly. 00:17:01.280 --> 00:17:03.440 align:middle line:84% There are a few things in Mandarin 00:17:03.440 --> 00:17:06.060 align:middle line:84% that people argue about whether they might be suffixes or not. 00:17:06.060 --> 00:17:08.900 align:middle line:84% So like you can make pronouns plural 00:17:08.900 --> 00:17:11.000 align:middle line:90% by adding a "men" after them. 00:17:11.000 --> 00:17:14.510 align:middle line:84% So "wo" is "I" and "women" is "we." 00:17:14.510 --> 00:17:17.329 align:middle line:84% So you add the "men" and you get a plural version. 00:17:17.329 --> 00:17:19.540 align:middle line:84% It's a candidate for a bound morpheme, 00:17:19.540 --> 00:17:21.290 align:middle line:84% but it's certainly a language that doesn't 00:17:21.290 --> 00:17:23.690 align:middle line:90% have a lot of bound morphemes. 00:17:23.690 --> 00:17:29.030 align:middle line:84% Opposite extreme are languages like Mohawk or Wampanoag. 00:17:29.030 --> 00:17:31.100 align:middle line:84% This is the language that is spoken 00:17:31.100 --> 00:17:34.170 align:middle line:84% by the traditional owners of a lot of Eastern Massachusetts, 00:17:34.170 --> 00:17:38.150 align:middle line:84% including, arguably, the place where we're standing right now. 00:17:38.150 --> 00:17:40.160 align:middle line:84% These languages are called polysynthetic. 00:17:40.160 --> 00:17:42.780 align:middle line:84% They are really, really into bound morphemes. 00:17:42.780 --> 00:17:46.580 align:middle line:84% So things that would not be bound morphemes in English 00:17:46.580 --> 00:17:47.690 align:middle line:90% are bound morphemes. 00:17:47.690 --> 00:17:52.370 align:middle line:84% This is an example from a letter that was written in the 1600s 00:17:52.370 --> 00:17:57.830 align:middle line:84% by a missionary to the Wampanoag on Martha's Vineyard, 00:17:57.830 --> 00:18:01.370 align:middle line:84% a guy named Experience Mayhew, because they knew 00:18:01.370 --> 00:18:03.590 align:middle line:90% how to give names in the 16th-- 00:18:03.590 --> 00:18:05.330 align:middle line:90% 1600s. 00:18:05.330 --> 00:18:07.430 align:middle line:84% He was asked, apparently, by the person 00:18:07.430 --> 00:18:09.260 align:middle line:84% he was writing the letter to to give him 00:18:09.260 --> 00:18:11.720 align:middle line:84% the longest word that he could in Wampanoag. 00:18:11.720 --> 00:18:14.600 align:middle line:84% And this is what he came up with. 00:18:14.600 --> 00:18:17.660 align:middle line:84% It means "our very skillful mirror makers." 00:18:17.660 --> 00:18:20.298 align:middle line:84% And as you can see, it consists of lots and lots of morphemes 00:18:20.298 --> 00:18:21.590 align:middle line:90% all piled on top of each other. 00:18:21.590 --> 00:18:24.510 align:middle line:84% He kind of cheated by putting "mirror" in the middle of that. 00:18:24.510 --> 00:18:27.920 align:middle line:84% So "mirror" itself is a morphologically complex word 00:18:27.920 --> 00:18:28.520 align:middle line:90% in Wampanoag. 00:18:28.520 --> 00:18:31.700 align:middle line:84% It means "device for looking at your reflection." 00:18:31.700 --> 00:18:36.320 align:middle line:84% Yeah, so look, "reflection device," [NON-ENGLISH],, 00:18:36.320 --> 00:18:38.085 align:middle line:84% is the word by itself for "mirror." 00:18:38.085 --> 00:18:39.710 align:middle line:84% So he sort of gave himself a head start 00:18:39.710 --> 00:18:40.970 align:middle line:84% by putting that in the middle of the word. 00:18:40.970 --> 00:18:41.470 align:middle line:90% Yeah. 00:18:41.470 --> 00:18:44.330 align:middle line:84% AUDIENCE: Is there any sort of connection 00:18:44.330 --> 00:18:46.910 align:middle line:84% between, I don't know, geography of where the language is 00:18:46.910 --> 00:18:50.710 align:middle line:84% spoken versus whether it's the language tends to be isolating? 00:18:50.710 --> 00:18:53.750 align:middle line:84% NORVIN RICHARDS: Which of these things it is? 00:18:53.750 --> 00:18:56.570 align:middle line:84% So the question was whether there's 00:18:56.570 --> 00:18:59.840 align:middle line:84% any connection between geography and whether a language 00:18:59.840 --> 00:19:01.850 align:middle line:90% is isolating or polysynthetic. 00:19:01.850 --> 00:19:05.580 align:middle line:84% These things kind of run in families, in language families. 00:19:05.580 --> 00:19:09.530 align:middle line:84% So there are a bunch of isolating Polynesian languages, 00:19:09.530 --> 00:19:10.970 align:middle line:90% for example. 00:19:10.970 --> 00:19:12.830 align:middle line:84% There are a bunch of polysynthetic. 00:19:12.830 --> 00:19:14.960 align:middle line:90% Wampanoag is Algonquian. 00:19:14.960 --> 00:19:16.430 align:middle line:90% Mohawk is Iroquoian. 00:19:16.430 --> 00:19:19.940 align:middle line:84% Those language families tend to be polysynthetic. 00:19:19.940 --> 00:19:23.600 align:middle line:84% And so because language families are 00:19:23.600 --> 00:19:25.370 align:middle line:84% confined to particular places, there's 00:19:25.370 --> 00:19:27.770 align:middle line:84% some kind of connection to where you are. 00:19:27.770 --> 00:19:30.620 align:middle line:84% I think that might be as far as it goes, actually. 00:19:30.620 --> 00:19:31.120 align:middle line:90% Yes. 00:19:31.120 --> 00:19:32.810 align:middle line:84% AUDIENCE: Does "polysynthetic" refer 00:19:32.810 --> 00:19:36.038 align:middle line:84% to the same group of languages as "agglutinative"? 00:19:36.038 --> 00:19:36.950 align:middle line:90% NORVIN RICHARDS: No. 00:19:36.950 --> 00:19:38.690 align:middle line:84% "Agglutinative" means something slightly different. 00:19:38.690 --> 00:19:40.607 align:middle line:84% We'll get to "agglutinative" in just a second. 00:19:40.607 --> 00:19:43.020 align:middle line:84% Yeah, that's a very good question. 00:19:43.020 --> 00:19:45.580 align:middle line:90% Other questions you want to ask? 00:19:45.580 --> 00:19:47.800 align:middle line:90% OK. 00:19:47.800 --> 00:19:49.930 align:middle line:84% I think we have "agglutinative" coming up. 00:19:49.930 --> 00:19:51.130 align:middle line:90% Yeah. 00:19:51.130 --> 00:19:53.470 align:middle line:84% So this is another, again, informal way 00:19:53.470 --> 00:19:55.870 align:middle line:90% of classifying languages. 00:19:55.870 --> 00:19:58.360 align:middle line:84% If we were having tests in this class, 00:19:58.360 --> 00:20:00.790 align:middle line:84% I would never test you on the meanings of these terms 00:20:00.790 --> 00:20:03.250 align:middle line:84% because they're fairly imprecise. 00:20:03.250 --> 00:20:06.340 align:middle line:84% But they're useful for giving people a basic idea of what 00:20:06.340 --> 00:20:07.750 align:middle line:90% you're looking at. 00:20:07.750 --> 00:20:11.050 align:middle line:84% Agglutinative language is-- this is another way people sometimes 00:20:11.050 --> 00:20:12.170 align:middle line:90% compare languages. 00:20:12.170 --> 00:20:13.810 align:middle line:84% Agglutinative languages are languages 00:20:13.810 --> 00:20:16.090 align:middle line:84% in which there are morphemes-- they're typically 00:20:16.090 --> 00:20:17.620 align:middle line:84% a bunch of morphemes that are bound, 00:20:17.620 --> 00:20:21.440 align:middle line:84% and they are easy to separate from each other. 00:20:21.440 --> 00:20:25.480 align:middle line:84% So Turkish is the poster child for an agglutinative language. 00:20:25.480 --> 00:20:28.240 align:middle line:84% You've got a verb, and then you have a bunch 00:20:28.240 --> 00:20:29.740 align:middle line:90% of suffixes on the verb. 00:20:29.740 --> 00:20:34.270 align:middle line:84% And making the cut between one suffix and the next 00:20:34.270 --> 00:20:36.050 align:middle line:90% is pretty straightforward. 00:20:36.050 --> 00:20:40.060 align:middle line:84% There aren't a lot of fancy morphological changes 00:20:40.060 --> 00:20:42.640 align:middle line:84% that happen as a result of these morphemes coming 00:20:42.640 --> 00:20:44.500 align:middle line:90% in contact with each other. 00:20:44.500 --> 00:20:47.740 align:middle line:84% And each of the morphemes has a pretty simple meaning. 00:20:47.740 --> 00:20:51.310 align:middle line:84% They mean things like, each other, or cause, or past. 00:20:51.310 --> 00:20:52.100 align:middle line:90% Yeah. 00:20:52.100 --> 00:20:52.600 align:middle line:90% Yes. 00:20:52.600 --> 00:20:55.540 align:middle line:84% AUDIENCE: So allomorphy wouldn't be the present [INAUDIBLE].. 00:20:55.540 --> 00:20:57.250 align:middle line:84% NORVIN RICHARDS: So, not in-- yeah, 00:20:57.250 --> 00:20:59.230 align:middle line:84% not the kind of allomorphy that causes 00:20:59.230 --> 00:21:03.220 align:middle line:84% you to wonder where one morpheme starts and the next begins. 00:21:03.220 --> 00:21:05.055 align:middle line:90% So there are languages-- 00:21:05.055 --> 00:21:07.180 align:middle line:84% the polysyn-- so there are polysynthetic languages. 00:21:07.180 --> 00:21:09.580 align:middle line:84% There are languages that have lots and lots of morphology 00:21:09.580 --> 00:21:10.390 align:middle line:90% on their verbs. 00:21:10.390 --> 00:21:11.980 align:middle line:90% Navajo is a very good example. 00:21:11.980 --> 00:21:13.870 align:middle line:84% We'll see some Navajo examples later, 00:21:13.870 --> 00:21:15.820 align:middle line:84% in which there are many morphemes 00:21:15.820 --> 00:21:17.860 align:middle line:84% but they tend to squash into each other. 00:21:17.860 --> 00:21:19.360 align:middle line:84% And you have to do a lot of analysis 00:21:19.360 --> 00:21:22.610 align:middle line:84% to figure out where one starts and the next begins-- 00:21:22.610 --> 00:21:24.230 align:middle line:84% where one stops and the next begins. 00:21:24.230 --> 00:21:24.830 align:middle line:90% Did you have a question? 00:21:24.830 --> 00:21:25.445 align:middle line:90% AUDIENCE: No. 00:21:25.445 --> 00:21:27.630 align:middle line:90% NORVIN RICHARDS: Yeah, OK. 00:21:27.630 --> 00:21:29.790 align:middle line:90% Further questions? 00:21:29.790 --> 00:21:30.950 align:middle line:90% OK. 00:21:30.950 --> 00:21:35.600 align:middle line:84% So agglutinative and sort of the opposite 00:21:35.600 --> 00:21:37.220 align:middle line:84% of agglutinative, people talk about 00:21:37.220 --> 00:21:39.290 align:middle line:84% "fusional," or "inflectional" languages. 00:21:39.290 --> 00:21:41.870 align:middle line:84% These are languages that have morphemes, 00:21:41.870 --> 00:21:45.590 align:middle line:84% and the morphemes squash a bunch of bits of meaning together 00:21:45.590 --> 00:21:46.920 align:middle line:90% in one small space. 00:21:46.920 --> 00:21:50.150 align:middle line:84% So these Russian nouns, for example, 00:21:50.150 --> 00:21:55.340 align:middle line:84% end in vowels, which indicates the grammatical gender 00:21:55.340 --> 00:21:59.570 align:middle line:84% of the noun, and also its number, and also its case. 00:21:59.570 --> 00:22:03.710 align:middle line:84% And you aren't going to be able to divide 00:22:03.710 --> 00:22:05.360 align:middle line:90% that vowel into parts, right? 00:22:05.360 --> 00:22:08.680 align:middle line:84% So the "oo" at the end of [NON-ENGLISH] means feminine 00:22:08.680 --> 00:22:10.040 align:middle line:90% and singular and accusative. 00:22:10.040 --> 00:22:13.010 align:middle line:84% And the "uh" at the end of [NON-ENGLISH] means feminine 00:22:13.010 --> 00:22:14.330 align:middle line:90% and plural and accusative. 00:22:14.330 --> 00:22:17.300 align:middle line:84% And it's hopeless to try to find a part of that suffix that's 00:22:17.300 --> 00:22:20.270 align:middle line:84% the part that means singular or plural, right? 00:22:20.270 --> 00:22:22.180 align:middle line:90% It's just all in one big spot. 00:22:22.180 --> 00:22:22.680 align:middle line:90% Yeah. 00:22:22.680 --> 00:22:25.008 align:middle line:84% AUDIENCE: Is this the case with endings in Spanish? 00:22:25.008 --> 00:22:26.300 align:middle line:90% NORVIN RICHARDS: Say it, again. 00:22:26.300 --> 00:22:27.740 align:middle line:84% AUDIENCE: Is this the case with endings in Spanish? 00:22:27.740 --> 00:22:29.360 align:middle line:90% NORVIN RICHARDS: Oh, I see. 00:22:29.360 --> 00:22:31.445 align:middle line:84% You're talking about the verb endings in Spanish? 00:22:31.445 --> 00:22:33.415 align:middle line:90% AUDIENCE: More like [INAUDIBLE]. 00:22:33.415 --> 00:22:35.870 align:middle line:90% NORVIN RICHARDS: Oh, I see. 00:22:35.870 --> 00:22:38.900 align:middle line:84% So this is why, I think I said-- so she's asking, 00:22:38.900 --> 00:22:43.126 align:middle line:84% is this true for Spanish nouns which end in endings like "-os" 00:22:43.126 --> 00:22:48.320 align:middle line:84% and "-as," right, which indicate both gender and number 00:22:48.320 --> 00:22:49.340 align:middle line:90% sometimes. 00:22:49.340 --> 00:22:50.960 align:middle line:84% And in those particular cases, you 00:22:50.960 --> 00:22:54.380 align:middle line:84% could try to convince yourself that the "oh" indicates 00:22:54.380 --> 00:22:57.440 align:middle line:84% masculine, and the S indicates plural, right? 00:22:57.440 --> 00:23:00.290 align:middle line:84% In lots of cases, this is why I said 00:23:00.290 --> 00:23:04.760 align:middle line:84% we're never going to test you on this kind of thing 00:23:04.760 --> 00:23:08.000 align:middle line:84% because this classification is really a classification of how 00:23:08.000 --> 00:23:10.670 align:middle line:84% easy is it to make divisions between morphemes 00:23:10.670 --> 00:23:12.930 align:middle line:90% and make breaks in places? 00:23:12.930 --> 00:23:16.190 align:middle line:84% And the answer is, sometimes very easy-- that's Turkish. 00:23:16.190 --> 00:23:18.380 align:middle line:84% Sometimes impossible, that's Russian. 00:23:18.380 --> 00:23:21.080 align:middle line:84% And Spanish is somewhere in the middle. 00:23:21.080 --> 00:23:23.420 align:middle line:84% It's more maybe for that particular case, 00:23:23.420 --> 00:23:24.920 align:middle line:84% it's more on the agglutinative side. 00:23:24.920 --> 00:23:26.587 align:middle line:84% Maybe there's a break that you can make. 00:23:26.587 --> 00:23:30.890 align:middle line:84% But, yeah, so asking questions like, how easy is it? 00:23:30.890 --> 00:23:32.270 align:middle line:90% This is no way to do science. 00:23:32.270 --> 00:23:33.163 align:middle line:90% Yeah. 00:23:33.163 --> 00:23:33.830 align:middle line:90% Other questions? 00:23:33.830 --> 00:23:34.330 align:middle line:90% Yes. 00:23:34.330 --> 00:23:35.614 align:middle line:90% AUDIENCE: [INAUDIBLE]. 00:23:35.614 --> 00:23:37.850 align:middle line:90% NORVIN RICHARDS: Oh, animate. 00:23:37.850 --> 00:23:38.660 align:middle line:90% Where? 00:23:38.660 --> 00:23:39.830 align:middle line:90% Oh, here. 00:23:39.830 --> 00:23:46.940 align:middle line:84% It stands for the fact that Russian masculine accusatives 00:23:46.940 --> 00:23:52.470 align:middle line:84% take "ah" only if they refer to things that are alive. 00:23:52.470 --> 00:23:52.970 align:middle line:90% Right. 00:23:52.970 --> 00:23:56.030 align:middle line:90% So, yeah, that's what it means. 00:23:56.030 --> 00:23:56.600 align:middle line:90% Yeah. 00:23:56.600 --> 00:24:00.470 align:middle line:84% So there's syncretism in Russian between the accusative 00:24:00.470 --> 00:24:04.100 align:middle line:84% and the genitive, I guess it is, just for animate nouns 00:24:04.100 --> 00:24:07.220 align:middle line:84% and not for inanimate masculines. 00:24:07.220 --> 00:24:08.360 align:middle line:90% Yeah. 00:24:08.360 --> 00:24:10.290 align:middle line:90% Any questions about this? 00:24:10.290 --> 00:24:12.110 align:middle line:90% OK. 00:24:12.110 --> 00:24:14.370 align:middle line:90% All right. 00:24:14.370 --> 00:24:14.870 align:middle line:90% OK. 00:24:14.870 --> 00:24:16.865 align:middle line:84% So I'm talking about lots of things 00:24:16.865 --> 00:24:17.990 align:middle line:90% that are language specific. 00:24:17.990 --> 00:24:20.090 align:middle line:90% And I just wanted to-- 00:24:20.090 --> 00:24:23.240 align:middle line:84% this is going to be a topic of conversation 00:24:23.240 --> 00:24:24.470 align:middle line:90% throughout the class. 00:24:24.470 --> 00:24:27.187 align:middle line:84% When we are looking at things, how much time 00:24:27.187 --> 00:24:29.270 align:middle line:84% are we spending looking at ways in which languages 00:24:29.270 --> 00:24:31.010 align:middle line:84% are different, and how much time are we 00:24:31.010 --> 00:24:33.770 align:middle line:84% spending trying to figure out the basic rules, 00:24:33.770 --> 00:24:36.270 align:middle line:84% the things that all languages have in common? 00:24:36.270 --> 00:24:39.650 align:middle line:84% Both of these are important things for linguists to do 00:24:39.650 --> 00:24:44.390 align:middle line:84% is try to understand both the variation and the core 00:24:44.390 --> 00:24:47.090 align:middle line:90% of common thread that all-- 00:24:47.090 --> 00:24:49.570 align:middle line:84% connects all these languages together. 00:24:49.570 --> 00:24:51.440 align:middle line:84% I just wanted to try to convince you 00:24:51.440 --> 00:24:53.960 align:middle line:84% that even in morphology, which is what we're starting with, 00:24:53.960 --> 00:24:59.930 align:middle line:84% there are some things that maybe are reliable 00:24:59.930 --> 00:25:01.430 align:middle line:90% cross-linguistically. 00:25:01.430 --> 00:25:05.240 align:middle line:84% So here's a Georgian verb and a Turkish verb. 00:25:05.240 --> 00:25:07.650 align:middle line:84% I don't know either of these languages 00:25:07.650 --> 00:25:09.650 align:middle line:84% so I had to pick examples that had 00:25:09.650 --> 00:25:11.250 align:middle line:90% a different verb at the bottom. 00:25:11.250 --> 00:25:13.310 align:middle line:84% So in Georgian, you've got "paint." 00:25:13.310 --> 00:25:15.440 align:middle line:84% And in Turkish, you've got "open." 00:25:15.440 --> 00:25:17.520 align:middle line:84% But otherwise, these verbs are the same. 00:25:17.520 --> 00:25:20.150 align:middle line:84% So the Georgian one means, "I will have him paint it." 00:25:20.150 --> 00:25:22.740 align:middle line:84% And the Turkish one means, "I had him open it." 00:25:22.740 --> 00:25:23.240 align:middle line:90% OK. 00:25:23.240 --> 00:25:23.780 align:middle line:90% They aren't the same. 00:25:23.780 --> 00:25:24.950 align:middle line:90% They're different tenses. 00:25:24.950 --> 00:25:27.252 align:middle line:84% But they're both about causing someone 00:25:27.252 --> 00:25:28.460 align:middle line:90% to do something to something. 00:25:28.460 --> 00:25:30.857 align:middle line:90% There, they have that in common. 00:25:30.857 --> 00:25:32.190 align:middle line:90% And they have some other things. 00:25:32.190 --> 00:25:33.898 align:middle line:84% And they're different in other ways, too. 00:25:33.898 --> 00:25:37.790 align:middle line:84% So the Georgian verb is preceded by several prefixes. 00:25:37.790 --> 00:25:42.350 align:middle line:84% The Turkish verb is followed by several suffixes. 00:25:42.350 --> 00:25:45.140 align:middle line:84% But here's what they finally have in common. 00:25:45.140 --> 00:25:51.830 align:middle line:84% The red affix is merged after the blue affix. 00:25:51.830 --> 00:25:56.720 align:middle line:84% So the affix that means "cause" is the first thing 00:25:56.720 --> 00:25:59.420 align:middle line:84% that's merged with the verb in both of these languages. 00:25:59.420 --> 00:26:01.820 align:middle line:84% And the affix that means first person 00:26:01.820 --> 00:26:06.890 align:middle line:84% singular, "I," as the subject, is merged later 00:26:06.890 --> 00:26:08.060 align:middle line:90% in both of these. 00:26:08.060 --> 00:26:10.820 align:middle line:84% Now, in Georgian, they're both prefixes. 00:26:10.820 --> 00:26:12.470 align:middle line:84% In Turkish, they're both suffixes. 00:26:12.470 --> 00:26:16.700 align:middle line:84% They're not pronounced the same way, right? 00:26:16.700 --> 00:26:19.100 align:middle line:90% "Cause" in Georgian is "a." 00:26:19.100 --> 00:26:21.710 align:middle line:90% "Cause" in Turkish is "tir." 00:26:21.710 --> 00:26:24.350 align:middle line:84% But they do have that property in common, 00:26:24.350 --> 00:26:28.070 align:middle line:84% that the order in which you combine these morphemes 00:26:28.070 --> 00:26:31.310 align:middle line:84% is the same for these two otherwise fairly different 00:26:31.310 --> 00:26:32.630 align:middle line:90% languages. 00:26:32.630 --> 00:26:35.000 align:middle line:84% And that's a pretty reliable thing, 00:26:35.000 --> 00:26:38.510 align:middle line:84% so what's sometimes called derivation morphology, which 00:26:38.510 --> 00:26:40.528 align:middle line:84% includes things like all the morphology 00:26:40.528 --> 00:26:42.320 align:middle line:84% we were talking about, that converts things 00:26:42.320 --> 00:26:45.080 align:middle line:84% from nouns into verbs or from verbs into adjectives 00:26:45.080 --> 00:26:48.440 align:middle line:84% or whatever all else, and also things 00:26:48.440 --> 00:26:51.950 align:middle line:84% like causative, that change the number of participants 00:26:51.950 --> 00:26:54.920 align:middle line:84% in a verb, so by adding someone who 00:26:54.920 --> 00:26:57.650 align:middle line:90% is causing a verb to happen. 00:26:57.650 --> 00:27:02.827 align:middle line:84% That type of morphology tends to be merged earlier than what's 00:27:02.827 --> 00:27:04.910 align:middle line:84% sometimes called inflectional morphology, the kind 00:27:04.910 --> 00:27:07.010 align:middle line:84% of morphology that indicates things like who's 00:27:07.010 --> 00:27:09.830 align:middle line:90% doing what to whom and when. 00:27:09.830 --> 00:27:12.245 align:middle line:84% That's a moderately reliable fact. 00:27:12.245 --> 00:27:15.110 align:middle line:90% 00:27:15.110 --> 00:27:18.050 align:middle line:84% Or similarly, there are various places 00:27:18.050 --> 00:27:21.680 align:middle line:84% where a pair of expressions can differ 00:27:21.680 --> 00:27:24.380 align:middle line:84% in all kinds of ways, including whether you're looking 00:27:24.380 --> 00:27:25.710 align:middle line:90% at prefixes or suffixes. 00:27:25.710 --> 00:27:28.465 align:middle line:84% So here's Swahili for "you hit" and German for "you put." 00:27:28.465 --> 00:27:30.590 align:middle line:84% And abstracting away from the fact that one of them 00:27:30.590 --> 00:27:33.240 align:middle line:84% is hitting and the other is putting, in Swahili, 00:27:33.240 --> 00:27:35.240 align:middle line:84% those are prefixes, in German, they're suffixes. 00:27:35.240 --> 00:27:37.220 align:middle line:84% But the order in which everything is merged 00:27:37.220 --> 00:27:39.350 align:middle line:90% is the same. 00:27:39.350 --> 00:27:42.860 align:middle line:84% In English and Turkish, "in my hands," and the Turkish version 00:27:42.860 --> 00:27:49.730 align:middle line:84% of that, "ellerimde," the Turkish morphemes are bound, 00:27:49.730 --> 00:27:51.170 align:middle line:90% the English morphemes are free. 00:27:51.170 --> 00:27:54.330 align:middle line:84% But the morphemes are being merged in the same order. 00:27:54.330 --> 00:27:56.120 align:middle line:84% So there's some hope for the idea 00:27:56.120 --> 00:27:59.450 align:middle line:84% that there is some core that these languages all share 00:27:59.450 --> 00:28:01.337 align:middle line:84% that we want to try to understand what 00:28:01.337 --> 00:28:03.170 align:middle line:84% is it that determines the order in which you 00:28:03.170 --> 00:28:04.533 align:middle line:90% do your merge operations. 00:28:04.533 --> 00:28:06.950 align:middle line:84% If you thought I was going to answer that question for you 00:28:06.950 --> 00:28:08.870 align:middle line:84% today, then I'm about to disappoint you. 00:28:08.870 --> 00:28:11.080 align:middle line:84% But that's a question that we're going to want 00:28:11.080 --> 00:28:12.330 align:middle line:90% to keep in mind going forward. 00:28:12.330 --> 00:28:13.050 align:middle line:90% Yeah? 00:28:13.050 --> 00:28:15.258 align:middle line:84% AUDIENCE: Does this have anything to do with the fact 00:28:15.258 --> 00:28:17.480 align:middle line:84% that, at least in English, there's like a certain way 00:28:17.480 --> 00:28:18.980 align:middle line:84% that you can say-- like, if you have 00:28:18.980 --> 00:28:20.780 align:middle line:84% a list of adjectives describing a noun, 00:28:20.780 --> 00:28:22.100 align:middle line:90% you say them in a certain way. 00:28:22.100 --> 00:28:25.340 align:middle line:84% So you say like "the big bad dog," 00:28:25.340 --> 00:28:27.170 align:middle line:90% but not the "bad big dog?" 00:28:27.170 --> 00:28:28.378 align:middle line:90% NORVIN RICHARDS: Bad big dog. 00:28:28.378 --> 00:28:31.242 align:middle line:84% Yeah, so that's a very nice point. 00:28:31.242 --> 00:28:33.200 align:middle line:84% When you have a bunch of adjectives in English, 00:28:33.200 --> 00:28:34.783 align:middle line:84% they tend to go in a particular order. 00:28:34.783 --> 00:28:36.260 align:middle line:90% That's right. 00:28:36.260 --> 00:28:39.440 align:middle line:84% And actually, there's a very articulated order. 00:28:39.440 --> 00:28:43.280 align:middle line:84% It's got like five slots for the adjectives and the order 00:28:43.280 --> 00:28:44.690 align:middle line:90% that they go in. 00:28:44.690 --> 00:28:47.210 align:middle line:84% The other cool thing about it is that 00:28:47.210 --> 00:28:51.690 align:middle line:84% among languages that put their adjectives before the noun, 00:28:51.690 --> 00:28:55.400 align:middle line:84% if they have any rules for the order of the adjective, that's 00:28:55.400 --> 00:28:56.690 align:middle line:90% the rule. 00:28:56.690 --> 00:28:59.540 align:middle line:84% So that's the only kind of rule that there can be. 00:28:59.540 --> 00:29:03.830 align:middle line:84% For languages that put their adjectives after the noun, 00:29:03.830 --> 00:29:05.570 align:middle line:84% if they have any rules for the order, 00:29:05.570 --> 00:29:06.870 align:middle line:90% there are two kinds of rules. 00:29:06.870 --> 00:29:08.990 align:middle line:84% So if you number the adjectives that 00:29:08.990 --> 00:29:11.690 align:middle line:84% go before the noun in English, if you say they go in the order 00:29:11.690 --> 00:29:15.380 align:middle line:84% 1, 2, 3, 4, 5, if you're a language that 00:29:15.380 --> 00:29:17.870 align:middle line:84% puts your adjectives after your nouns, 00:29:17.870 --> 00:29:20.090 align:middle line:84% there are only two kinds of languages like that. 00:29:20.090 --> 00:29:22.820 align:middle line:84% There are languages in which the order is 1, 2, 3, 4, 5, 00:29:22.820 --> 00:29:24.740 align:middle line:90% so you get noun, 1, 2, 3, 4, 5. 00:29:24.740 --> 00:29:28.610 align:middle line:84% So it'd be like you say "house big black," 00:29:28.610 --> 00:29:31.340 align:middle line:84% and you can't say "house black big." 00:29:31.340 --> 00:29:37.640 align:middle line:84% You say, "A little green Japanese carving knife," right? 00:29:37.640 --> 00:29:42.530 align:middle line:84% And in these languages, you would say "A knife little green 00:29:42.530 --> 00:29:43.370 align:middle line:90% Japanese carving." 00:29:43.370 --> 00:29:45.620 align:middle line:84% So you have the adjectives in the same order they'd be 00:29:45.620 --> 00:29:47.060 align:middle line:90% in English, but after the noun. 00:29:47.060 --> 00:29:49.050 align:middle line:84% That's one kind of adjective-after-the-noun 00:29:49.050 --> 00:29:49.550 align:middle line:90% language. 00:29:49.550 --> 00:29:50.967 align:middle line:84% And the other kind is just the one 00:29:50.967 --> 00:29:53.730 align:middle line:84% that has the mirror image of the adjectives. 00:29:53.730 --> 00:29:55.940 align:middle line:84% So in the languages of the world, there are, I think, 00:29:55.940 --> 00:29:58.280 align:middle line:84% three kinds, which is kind of interesting. 00:29:58.280 --> 00:29:59.210 align:middle line:90% Yes? 00:29:59.210 --> 00:30:00.800 align:middle line:84% AUDIENCE: Why is it sort of right 00:30:00.800 --> 00:30:03.560 align:middle line:84% to say "big black house" but not "black big house?" 00:30:03.560 --> 00:30:06.120 align:middle line:84% NORVIN RICHARDS: Hey, good question, yes. 00:30:06.120 --> 00:30:07.290 align:middle line:90% Yeah. 00:30:07.290 --> 00:30:11.310 align:middle line:84% There are linguists working on that, yes. 00:30:11.310 --> 00:30:13.990 align:middle line:84% And that's actually basically her question, your question, 00:30:13.990 --> 00:30:15.240 align:middle line:90% from just a second ago, right? 00:30:15.240 --> 00:30:19.620 align:middle line:84% So there are conditions on the ordering of the adjectives. 00:30:19.620 --> 00:30:22.380 align:middle line:84% And yes, we want to try to understand what those are. 00:30:22.380 --> 00:30:24.120 align:middle line:84% All of you are eventually going to want 00:30:24.120 --> 00:30:26.035 align:middle line:84% to write a research proposal in which you 00:30:26.035 --> 00:30:28.410 align:middle line:84% try to tell me that you're going to figure something out. 00:30:28.410 --> 00:30:29.940 align:middle line:84% I can give you some things to read 00:30:29.940 --> 00:30:31.470 align:middle line:84% that are attempting to figure out 00:30:31.470 --> 00:30:32.760 align:middle line:90% the answer to that question. 00:30:32.760 --> 00:30:37.000 align:middle line:84% But yeah, that's a research topic. 00:30:37.000 --> 00:30:37.960 align:middle line:90% Other questions? 00:30:37.960 --> 00:30:40.540 align:middle line:90% 00:30:40.540 --> 00:30:41.370 align:middle line:90% Cool, all right. 00:30:41.370 --> 00:30:45.600 align:middle line:90% 00:30:45.600 --> 00:30:49.470 align:middle line:84% So yeah, this is just me saying again what I said. 00:30:49.470 --> 00:30:51.330 align:middle line:84% So what those trees have in common 00:30:51.330 --> 00:30:53.640 align:middle line:84% is that if A is higher than B in one tree, 00:30:53.640 --> 00:30:55.440 align:middle line:84% the same A is higher than the same B 00:30:55.440 --> 00:30:58.350 align:middle line:84% in the corresponding tree, where by 00:30:58.350 --> 00:31:01.680 align:middle line:84% "higher," we mean higher in the tree, 00:31:01.680 --> 00:31:05.310 align:middle line:84% added later in the series of merge operations. 00:31:05.310 --> 00:31:06.210 align:middle line:90% That's what we mean. 00:31:06.210 --> 00:31:08.820 align:middle line:90% 00:31:08.820 --> 00:31:11.580 align:middle line:84% Lots of questions, like the question that both of you 00:31:11.580 --> 00:31:12.300 align:middle line:90% asked-- 00:31:12.300 --> 00:31:14.610 align:middle line:84% what's going on with order of adjectives, 00:31:14.610 --> 00:31:16.740 align:middle line:84% or how come these things have to be 00:31:16.740 --> 00:31:19.290 align:middle line:84% merged in these orders in all of the trees that I'm showing. 00:31:19.290 --> 00:31:21.832 align:middle line:84% And I'm not offering to answer those questions for you today. 00:31:21.832 --> 00:31:24.330 align:middle line:84% These are some of the central questions of syntax, 00:31:24.330 --> 00:31:27.363 align:middle line:84% and we'll try to understand them eventually. 00:31:27.363 --> 00:31:31.400 align:middle line:90% OK, yeah? 00:31:31.400 --> 00:31:34.070 align:middle line:84% OK, so we started off by saying there 00:31:34.070 --> 00:31:37.498 align:middle line:84% are at least two imaginable kinds of lexicons if we're 00:31:37.498 --> 00:31:39.290 align:middle line:84% going to make a list of everything you know 00:31:39.290 --> 00:31:44.720 align:middle line:84% about your language, one that has in it entries like "teach," 00:31:44.720 --> 00:31:47.420 align:middle line:84% "teacher," "teachers," "teaching," and "mine," 00:31:47.420 --> 00:31:50.040 align:middle line:84% "miner," "miners," "mining," and so on, 00:31:50.040 --> 00:31:52.910 align:middle line:84% and another one that instead has-- 00:31:52.910 --> 00:31:57.330 align:middle line:84% so that first lexicon consists entirely of words. 00:31:57.330 --> 00:32:00.290 align:middle line:84% If you look at an actual dictionary, that's 00:32:00.290 --> 00:32:02.450 align:middle line:84% the kind of lexicon you're probably used to seeing. 00:32:02.450 --> 00:32:03.950 align:middle line:84% But we've been fooling with the idea 00:32:03.950 --> 00:32:06.950 align:middle line:84% that your mental lexicon is not organized that way. 00:32:06.950 --> 00:32:09.300 align:middle line:84% What your mental lexicon has in it is morphemes. 00:32:09.300 --> 00:32:12.830 align:middle line:84% So it has "teach," and "mine," and suffixes like "-er" 00:32:12.830 --> 00:32:17.210 align:middle line:84% and "-ing," and rules for how morphemes can be combined with 00:32:17.210 --> 00:32:18.950 align:middle line:84% each other, and what the result is, 00:32:18.950 --> 00:32:21.750 align:middle line:84% and what happens when morphemes combine with each other. 00:32:21.750 --> 00:32:24.320 align:middle line:84% That's the kind of model of how language is put together 00:32:24.320 --> 00:32:25.880 align:middle line:84% that we've been playing around with. 00:32:25.880 --> 00:32:27.240 align:middle line:90% Yeah? 00:32:27.240 --> 00:32:28.940 align:middle line:90% OK. 00:32:28.940 --> 00:32:30.410 align:middle line:90% All right. 00:32:30.410 --> 00:32:33.620 align:middle line:84% Yes, so just to give these theories 00:32:33.620 --> 00:32:35.960 align:middle line:84% names, nice nonjudgmental names, we're 00:32:35.960 --> 00:32:39.350 align:middle line:84% going to call them the wrong theory and the right theory. 00:32:39.350 --> 00:32:43.910 align:middle line:84% And evidence that the right theory is right, 00:32:43.910 --> 00:32:47.210 align:middle line:90% that we played around with-- 00:32:47.210 --> 00:32:49.160 align:middle line:84% for one thing, I asked you to take pity 00:32:49.160 --> 00:32:50.600 align:middle line:90% on the plight of the Nimborans. 00:32:50.600 --> 00:32:53.240 align:middle line:84% These are these people in Papua New Guinea whose verbs each 00:32:53.240 --> 00:32:55.970 align:middle line:90% have 27,000 forms. 00:32:55.970 --> 00:32:59.360 align:middle line:84% If we make their lexicons have every verb form in it, 00:32:59.360 --> 00:33:01.250 align:middle line:84% their lexicons are going to be very large. 00:33:01.250 --> 00:33:04.040 align:middle line:84% They will have no time to do anything other than sit around 00:33:04.040 --> 00:33:07.340 align:middle line:84% and remember verbs, whereas if we're 00:33:07.340 --> 00:33:11.300 align:middle line:84% willing to allow them to break words into morphemes 00:33:11.300 --> 00:33:13.070 align:middle line:84% and put those in their lexicons, then 00:33:13.070 --> 00:33:15.860 align:middle line:84% they will have lexicons of a reasonable size. 00:33:15.860 --> 00:33:19.490 align:middle line:84% And we saw lots of evidence that people manipulate morphemes, 00:33:19.490 --> 00:33:24.270 align:middle line:84% including the fact that you can apply morphemes to new words. 00:33:24.270 --> 00:33:27.848 align:middle line:84% So if I tell you I've invented a new thing, it's called a "wug," 00:33:27.848 --> 00:33:29.390 align:middle line:84% and then I show you another one, then 00:33:29.390 --> 00:33:31.740 align:middle line:84% you will call the two things "wugs." 00:33:31.740 --> 00:33:34.927 align:middle line:84% This is a classic experiment by Jean Berko Gleason, which 00:33:34.927 --> 00:33:37.010 align:middle line:84% she did this with small children and they did what 00:33:37.010 --> 00:33:39.030 align:middle line:90% you would expect them to do-- 00:33:39.030 --> 00:33:40.610 align:middle line:90% and so on. 00:33:40.610 --> 00:33:41.190 align:middle line:90% Yeah? 00:33:41.190 --> 00:33:41.690 align:middle line:90% Good? 00:33:41.690 --> 00:33:44.060 align:middle line:84% People are willing to believe in morphemes? 00:33:44.060 --> 00:33:45.990 align:middle line:90% Excellent. 00:33:45.990 --> 00:33:47.790 align:middle line:90% OK. 00:33:47.790 --> 00:33:52.380 align:middle line:90% I want to spend some time, then. 00:33:52.380 --> 00:33:55.650 align:middle line:84% The problem set that is now finally available to you 00:33:55.650 --> 00:33:58.390 align:middle line:90% has on it a morphology problem. 00:33:58.390 --> 00:34:01.290 align:middle line:84% So it gives you a bunch of data from a language. 00:34:01.290 --> 00:34:02.940 align:middle line:84% The language is Inupiaq, which is 00:34:02.940 --> 00:34:07.620 align:middle line:84% a language spoken in the Arctic, in Alaska, 00:34:07.620 --> 00:34:10.659 align:middle line:90% in the coast of Alaska. 00:34:10.659 --> 00:34:12.210 align:middle line:84% And it asks you to go through and try 00:34:12.210 --> 00:34:13.877 align:middle line:84% to find all the morphemes and figure out 00:34:13.877 --> 00:34:15.852 align:middle line:90% rules for how they combine. 00:34:15.852 --> 00:34:18.060 align:middle line:84% We've already done a little bit of that kind of thing 00:34:18.060 --> 00:34:22.020 align:middle line:84% in this class, but I wanted to do another problem like this. 00:34:22.020 --> 00:34:25.355 align:middle line:84% And this problem is going to introduce some new issues 00:34:25.355 --> 00:34:26.730 align:middle line:84% and things for you to think about 00:34:26.730 --> 00:34:27.855 align:middle line:90% as you're doing morphology. 00:34:27.855 --> 00:34:29.585 align:middle line:90% So I want to go through these. 00:34:29.585 --> 00:34:31.710 align:middle line:84% These data are from a language called Lardil, which 00:34:31.710 --> 00:34:32.960 align:middle line:90% I think I've mentioned before. 00:34:32.960 --> 00:34:35.050 align:middle line:84% It's a language spoken in northern Australia 00:34:35.050 --> 00:34:38.550 align:middle line:84% which I've done in field work on. 00:34:38.550 --> 00:34:42.750 align:middle line:84% It's a language that has inflection on its nouns. 00:34:42.750 --> 00:34:44.710 align:middle line:84% Its nouns get inflected for case. 00:34:44.710 --> 00:34:47.830 align:middle line:84% So they have suffixes indicating case. 00:34:47.830 --> 00:34:51.210 align:middle line:84% And if you're not sure what I meant by that, here's 00:34:51.210 --> 00:34:52.560 align:middle line:90% what I mean. 00:34:52.560 --> 00:34:55.080 align:middle line:84% Nouns have little markers on them telling basically 00:34:55.080 --> 00:34:57.940 align:middle line:84% whether they are the subject or the object of a verb. 00:34:57.940 --> 00:34:59.610 align:middle line:84% So if you're the subject of a verb, 00:34:59.610 --> 00:35:01.270 align:middle line:90% you're in the nominative case. 00:35:01.270 --> 00:35:05.680 align:middle line:84% And if you're the object, then you're in the accusative case. 00:35:05.680 --> 00:35:08.010 align:middle line:84% If you study-- English does not have so much of this, 00:35:08.010 --> 00:35:09.180 align:middle line:90% except on our pronouns. 00:35:09.180 --> 00:35:10.770 align:middle line:84% If you study lots of other languages, 00:35:10.770 --> 00:35:15.037 align:middle line:84% if you try to learn Russian, or Japanese, or Latin, 00:35:15.037 --> 00:35:17.370 align:middle line:84% or many other languages, you've had to learn about case. 00:35:17.370 --> 00:35:18.630 align:middle line:90% Yeah? 00:35:18.630 --> 00:35:21.210 align:middle line:90% So Lardil has case. 00:35:21.210 --> 00:35:24.810 align:middle line:84% Here are a bunch of nouns in their nominative and accusative 00:35:24.810 --> 00:35:25.620 align:middle line:90% forms. 00:35:25.620 --> 00:35:28.500 align:middle line:84% So you've got "mela," which means "seawater." 00:35:28.500 --> 00:35:30.690 align:middle line:84% It also means "beer," I guess on the theory 00:35:30.690 --> 00:35:33.510 align:middle line:84% that they're both foamy things that make you sick 00:35:33.510 --> 00:35:35.790 align:middle line:90% if you drink too much of it. 00:35:35.790 --> 00:35:39.810 align:middle line:84% "Mela," "barnga," "thungal," "ketharr," and "miyar," those 00:35:39.810 --> 00:35:42.030 align:middle line:84% are the Lardil nouns in their nominative forms. 00:35:42.030 --> 00:35:44.072 align:middle line:84% And then you've got their accusative forms there, 00:35:44.072 --> 00:35:46.080 align:middle line:90% which I won't read to you. 00:35:46.080 --> 00:35:48.944 align:middle line:90% What's the accusative morpheme? 00:35:48.944 --> 00:35:50.360 align:middle line:90% AUDIENCE: "-n." 00:35:50.360 --> 00:35:51.530 align:middle line:90% NORVIN RICHARDS: "-n." 00:35:51.530 --> 00:35:52.550 align:middle line:90% Joseph? 00:35:52.550 --> 00:35:54.347 align:middle line:90% "-n," yeah. 00:35:54.347 --> 00:35:56.930 align:middle line:84% Oh dear, I'm going to have to do this writing-everything-twice 00:35:56.930 --> 00:35:57.430 align:middle line:90% thing again. 00:35:57.430 --> 00:36:01.050 align:middle line:84% I really need to learn how to manipulate the screens better. 00:36:01.050 --> 00:36:04.760 align:middle line:84% So the accusative suffix is "-n." 00:36:04.760 --> 00:36:07.020 align:middle line:84% Say it again, the accusative suffix is "-n." 00:36:07.020 --> 00:36:11.820 align:middle line:84% Does anybody find any examples that are a problem for that? 00:36:11.820 --> 00:36:12.720 align:middle line:90% Yes? 00:36:12.720 --> 00:36:14.460 align:middle line:84% AUDIENCE: The ones that have "-in." 00:36:14.460 --> 00:36:16.835 align:middle line:84% NORVIN RICHARDS: Yeah, so some of them have "-in," right? 00:36:16.835 --> 00:36:19.830 align:middle line:84% Like "tree," and "river," and "spear," I think. 00:36:19.830 --> 00:36:23.190 align:middle line:84% What determines whether the suffix is "-n" or "-in"? 00:36:23.190 --> 00:36:24.237 align:middle line:90% Yes? 00:36:24.237 --> 00:36:25.320 align:middle line:90% AUDIENCE: The last letter. 00:36:25.320 --> 00:36:29.355 align:middle line:84% NORVIN RICHARDS: The last letter of the nominative, yep. 00:36:29.355 --> 00:36:30.470 align:middle line:90% AUDIENCE: "a" or not "a." 00:36:30.470 --> 00:36:32.220 align:middle line:84% NORVIN RICHARDS: Can you be more specific? 00:36:32.220 --> 00:36:35.460 align:middle line:84% What determines whether it's just an N or an I-N? 00:36:35.460 --> 00:36:38.633 align:middle line:84% AUDIENCE: If the noun ends in "a," then it ends with "-n." 00:36:38.633 --> 00:36:40.312 align:middle line:84% If it ends in not "a," then it's "-in." 00:36:40.312 --> 00:36:41.520 align:middle line:90% NORVIN RICHARDS: That's true. 00:36:41.520 --> 00:36:44.690 align:middle line:84% All of these examples, it ends in an "a." 00:36:44.690 --> 00:36:46.440 align:middle line:84% AUDIENCE: Are they vowel versus consonant? 00:36:46.440 --> 00:36:48.555 align:middle line:84% NORVIN RICHARDS: Yeah, so we might bravely 00:36:48.555 --> 00:36:50.430 align:middle line:84% say that it's about vowels versus consonants, 00:36:50.430 --> 00:36:51.060 align:middle line:90% and we'd be right. 00:36:51.060 --> 00:36:52.227 align:middle line:90% But you're absolutely right. 00:36:52.227 --> 00:36:54.120 align:middle line:90% I haven't shown you that. 00:36:54.120 --> 00:36:56.910 align:middle line:84% So the fact is, after a consonant, 00:36:56.910 --> 00:36:59.310 align:middle line:84% you get "-in," and after a vowel, you get "n." 00:36:59.310 --> 00:37:02.250 align:middle line:84% I'm going to say that again over here. 00:37:02.250 --> 00:37:05.310 align:middle line:90% After a consonant, you get-- 00:37:05.310 --> 00:37:07.410 align:middle line:84% sorry, after a vowel, you get "-n." 00:37:07.410 --> 00:37:09.870 align:middle line:84% And after a consonant, you get -in." 00:37:09.870 --> 00:37:11.110 align:middle line:90% OK? 00:37:11.110 --> 00:37:11.610 align:middle line:90% Good. 00:37:11.610 --> 00:37:12.450 align:middle line:90% So far, so good. 00:37:12.450 --> 00:37:15.200 align:middle line:90% 00:37:15.200 --> 00:37:16.370 align:middle line:90% Here are some more nouns. 00:37:16.370 --> 00:37:19.350 align:middle line:84% So the nouns appear, the nouns you saw before. 00:37:19.350 --> 00:37:20.840 align:middle line:90% There are some more nouns. 00:37:20.840 --> 00:37:24.830 align:middle line:84% "Rain" is when "wunda," "tip" is "belda," and "curve" 00:37:24.830 --> 00:37:26.150 align:middle line:90% is "dalda." 00:37:26.150 --> 00:37:30.080 align:middle line:84% And the accusatives are "wunin," "belin," and "dalin." 00:37:30.080 --> 00:37:34.325 align:middle line:84% That's unexpected, right, given our theory so far? 00:37:34.325 --> 00:37:39.900 align:middle line:90% 00:37:39.900 --> 00:37:42.140 align:middle line:90% What do you think is going on? 00:37:42.140 --> 00:37:42.770 align:middle line:90% Yes? 00:37:42.770 --> 00:37:46.807 align:middle line:84% AUDIENCE: If you remove the "-da" at the end, [INAUDIBLE].. 00:37:46.807 --> 00:37:48.890 align:middle line:84% NORVIN RICHARDS: So it's as though you're removing 00:37:48.890 --> 00:37:52.550 align:middle line:84% the "-da," right, at the end, yeah? 00:37:52.550 --> 00:37:53.501 align:middle line:90% Yes? 00:37:53.501 --> 00:37:56.150 align:middle line:90% AUDIENCE: What you think is-- 00:37:56.150 --> 00:38:01.100 align:middle line:84% I think it's more than just "-da," because you have "wunda" 00:38:01.100 --> 00:38:02.030 align:middle line:90% opposite to "katha." 00:38:02.030 --> 00:38:03.238 align:middle line:90% NORVIN RICHARDS: Yeah, we do. 00:38:03.238 --> 00:38:04.310 align:middle line:90% We have a minimal pair. 00:38:04.310 --> 00:38:06.210 align:middle line:84% It was sneaky of me to put that in there. 00:38:06.210 --> 00:38:10.350 align:middle line:84% So there's a "wunda" that means a type of stingray. 00:38:10.350 --> 00:38:11.990 align:middle line:84% The accusative of that is "wundan." 00:38:11.990 --> 00:38:14.397 align:middle line:84% And then there's another one down here that means rain, 00:38:14.397 --> 00:38:15.980 align:middle line:84% and the accusative of that is "wunin." 00:38:15.980 --> 00:38:18.580 align:middle line:90% 00:38:18.580 --> 00:38:22.210 align:middle line:84% So here I am asking you, how do we get from the nominative 00:38:22.210 --> 00:38:23.800 align:middle line:90% to the accusative? 00:38:23.800 --> 00:38:24.607 align:middle line:90% Yeah? 00:38:24.607 --> 00:38:26.940 align:middle line:84% AUDIENCE: Is "-da" a morpheme that turns everything else 00:38:26.940 --> 00:38:28.040 align:middle line:90% into the nominative? 00:38:28.040 --> 00:38:29.260 align:middle line:84% NORVIN RICHARDS: It would be nice, but no. 00:38:29.260 --> 00:38:31.510 align:middle line:84% Well, actually that's an interesting way to say it. 00:38:31.510 --> 00:38:34.160 align:middle line:84% There's a way to say what you just said that would be true. 00:38:34.160 --> 00:38:35.800 align:middle line:84% So maybe I shouldn't just say no. 00:38:35.800 --> 00:38:36.550 align:middle line:90% Yes? 00:38:36.550 --> 00:38:40.060 align:middle line:84% AUDIENCE: Is the verb actually something else, and they're 00:38:40.060 --> 00:38:43.660 align:middle line:84% both being modified to form the causative adjective? 00:38:43.660 --> 00:38:46.450 align:middle line:84% NORVIN RICHARDS: So I think you and you-- actually, I'm sorry. 00:38:46.450 --> 00:38:48.580 align:middle line:84% I shouldn't have spoken so quickly-- 00:38:48.580 --> 00:38:49.660 align:middle line:90% are on the right track. 00:38:49.660 --> 00:38:51.580 align:middle line:84% Look, here I am saying, how can we 00:38:51.580 --> 00:38:53.890 align:middle line:84% get from the nominative to the accusative, right? 00:38:53.890 --> 00:38:56.620 align:middle line:84% And the answer is despair, right? 00:38:56.620 --> 00:38:59.320 align:middle line:84% That's what that minimal pair is meant to do. 00:38:59.320 --> 00:39:01.090 align:middle line:84% We had the same thing in Polish, right? 00:39:01.090 --> 00:39:04.600 align:middle line:84% Where the word for "lye" and the word for something else 00:39:04.600 --> 00:39:07.948 align:middle line:84% were the same in the singular but different in the plural. 00:39:07.948 --> 00:39:09.490 align:middle line:84% So we convinced ourselves, all right, 00:39:09.490 --> 00:39:12.280 align:middle line:84% we can't predict the plural from the singular. 00:39:12.280 --> 00:39:14.320 align:middle line:84% But we can predict the singular from the plural. 00:39:14.320 --> 00:39:15.710 align:middle line:90% That's what we decided, right? 00:39:15.710 --> 00:39:18.760 align:middle line:84% So in Polish, we said some of these nouns underlyingly 00:39:18.760 --> 00:39:20.980 align:middle line:84% end in "g," some of them underlyingly end in "k." 00:39:20.980 --> 00:39:23.320 align:middle line:84% And there's a general rule that "g" becomes 00:39:23.320 --> 00:39:24.705 align:middle line:90% "k" at the end of a word. 00:39:24.705 --> 00:39:26.080 align:middle line:84% Can we do something similar here? 00:39:26.080 --> 00:39:27.650 align:middle line:90% Yes? 00:39:27.650 --> 00:39:29.760 align:middle line:84% AUDIENCE: Also, how do we know that "wunda" is 00:39:29.760 --> 00:39:32.002 align:middle line:90% "wunda," the stingray species? 00:39:32.002 --> 00:39:35.035 align:middle line:84% Is there some kind of exception to the rule? 00:39:35.035 --> 00:39:38.530 align:middle line:84% NORVIN RICHARDS: Could be that, could be that. 00:39:38.530 --> 00:39:40.570 align:middle line:90% Two things, it isn't. 00:39:40.570 --> 00:39:43.750 align:middle line:90% It's part of a general thing. 00:39:43.750 --> 00:39:45.670 align:middle line:90% But also, wait. 00:39:45.670 --> 00:39:46.840 align:middle line:90% You're absolutely right. 00:39:46.840 --> 00:39:48.880 align:middle line:84% Whenever we're looking at a data set, 00:39:48.880 --> 00:39:50.380 align:middle line:84% we could say to ourselves, well, you 00:39:50.380 --> 00:39:52.060 align:middle line:84% know what, maybe this is just an exception. 00:39:52.060 --> 00:39:53.870 align:middle line:84% Maybe I shouldn't try to account for this. 00:39:53.870 --> 00:39:59.280 align:middle line:84% But in this class, you should trust me not to do that to you. 00:39:59.280 --> 00:40:01.460 align:middle line:84% So I'm going to give you data sets in which you 00:40:01.460 --> 00:40:04.190 align:middle line:84% will be able to come up with rules that account for things. 00:40:04.190 --> 00:40:08.930 align:middle line:84% And in life, that shouldn't be your first move, probably, 00:40:08.930 --> 00:40:09.863 align:middle line:90% because-- 00:40:09.863 --> 00:40:11.030 align:middle line:90% I don't want to pick on you. 00:40:11.030 --> 00:40:13.310 align:middle line:84% It's a move you hear people make a lot. 00:40:13.310 --> 00:40:15.110 align:middle line:90% Hey, maybe this is an exception. 00:40:15.110 --> 00:40:17.240 align:middle line:84% But making that your first move is 00:40:17.240 --> 00:40:19.250 align:middle line:84% kind of a recipe for failing to make 00:40:19.250 --> 00:40:20.465 align:middle line:90% exciting discoveries, right? 00:40:20.465 --> 00:40:22.340 align:middle line:84% So if you're willing to exclude certain data, 00:40:22.340 --> 00:40:23.660 align:middle line:84% I'm just not going to account for that. 00:40:23.660 --> 00:40:25.520 align:middle line:84% Sometimes, that's the way to actually account 00:40:25.520 --> 00:40:27.520 align:middle line:84% for everything, is by being willing to recognize 00:40:27.520 --> 00:40:28.310 align:middle line:90% exceptions. 00:40:28.310 --> 00:40:30.770 align:middle line:84% But other times, what it does is stop you from realizing, 00:40:30.770 --> 00:40:33.530 align:middle line:84% no, the exceptions have a general character. 00:40:33.530 --> 00:40:36.050 align:middle line:84% We can understand what they all are. 00:40:36.050 --> 00:40:37.820 align:middle line:90% So it's worth like-- 00:40:37.820 --> 00:40:39.560 align:middle line:84% maybe that can be part of a strategy, 00:40:39.560 --> 00:40:41.420 align:middle line:84% corralling all the exceptions into one place 00:40:41.420 --> 00:40:43.045 align:middle line:84% and then looking at them all and trying 00:40:43.045 --> 00:40:45.302 align:middle line:90% to understand what's going on. 00:40:45.302 --> 00:40:46.760 align:middle line:84% In this particular case, the things 00:40:46.760 --> 00:40:49.460 align:middle line:84% that look exceptional right now are "rain," "tip," and "curve." 00:40:49.460 --> 00:40:51.860 align:middle line:84% We had this beautiful rule on the basis of the words 00:40:51.860 --> 00:40:52.910 align:middle line:90% that were up at the top. 00:40:52.910 --> 00:40:55.490 align:middle line:84% And now, these new data are breaking it, right? 00:40:55.490 --> 00:40:58.820 align:middle line:84% We want the accusative of "belda" to be "beldan." 00:40:58.820 --> 00:41:02.930 align:middle line:84% The fact that it's "belin" is weird. 00:41:02.930 --> 00:41:04.910 align:middle line:84% Is there a way to do the trick that we 00:41:04.910 --> 00:41:09.890 align:middle line:84% did for Polish, which was to say the plural is the useful-- 00:41:09.890 --> 00:41:12.765 align:middle line:84% is the thing that we can use to predict the singular, right? 00:41:12.765 --> 00:41:14.390 align:middle line:84% So we'll start with the plural and then 00:41:14.390 --> 00:41:16.140 align:middle line:84% we'll figure out how to make the singular. 00:41:16.140 --> 00:41:20.000 align:middle line:90% Can we play that game here? 00:41:20.000 --> 00:41:20.893 align:middle line:90% Yeah. 00:41:20.893 --> 00:41:22.310 align:middle line:84% AUDIENCE: Is there a case form now 00:41:22.310 --> 00:41:25.370 align:middle line:84% that we don't know, that is, you get something added onto that 00:41:25.370 --> 00:41:27.410 align:middle line:84% complement to nominative to form the accusative? 00:41:27.410 --> 00:41:29.960 align:middle line:84% NORVIN RICHARDS: So maybe that's a way to think about it. 00:41:29.960 --> 00:41:32.335 align:middle line:84% And I think this is more or less what several people have 00:41:32.335 --> 00:41:32.930 align:middle line:90% been saying. 00:41:32.930 --> 00:41:39.590 align:middle line:84% What do you wish the nominative of this was, given our rules? 00:41:39.590 --> 00:41:44.330 align:middle line:84% So we wish that they were "wun," "bel," and "dal," right? 00:41:44.330 --> 00:41:50.940 align:middle line:84% Yeah, so I'm just going to write that down, "wun," "bel," 00:41:50.940 --> 00:41:52.540 align:middle line:90% and "dal." 00:41:52.540 --> 00:41:54.600 align:middle line:84% And then if it were that, well, we 00:41:54.600 --> 00:41:56.795 align:middle line:84% would be able to get the accusative, 00:41:56.795 --> 00:41:58.170 align:middle line:84% and we would need to do something 00:41:58.170 --> 00:41:59.760 align:middle line:90% to get the nominative. 00:41:59.760 --> 00:42:00.310 align:middle line:90% Yeah, Vlada? 00:42:00.310 --> 00:42:02.310 align:middle line:84% AUDIENCE: I want to say something about the fact 00:42:02.310 --> 00:42:04.490 align:middle line:84% that all of them are one syllable. 00:42:04.490 --> 00:42:06.948 align:middle line:90% [INAUDIBLE] for whatever reason. 00:42:06.948 --> 00:42:09.240 align:middle line:84% NORVIN RICHARDS: I think that would be very attractive, 00:42:09.240 --> 00:42:11.435 align:middle line:84% so I think that is raising a good point. 00:42:11.435 --> 00:42:13.560 align:middle line:84% All of the other words that I've shown you, "mela," 00:42:13.560 --> 00:42:15.977 align:middle line:84% "barnga," "katha," "wunda," "thungal," "ketharr," "miyar," 00:42:15.977 --> 00:42:18.390 align:middle line:90% they're all two syllables long. 00:42:18.390 --> 00:42:22.140 align:middle line:84% These words, if there were words like "wun," "bel," and "dal," 00:42:22.140 --> 00:42:25.410 align:middle line:84% they would be our first Lardil monosyllables. 00:42:25.410 --> 00:42:27.270 align:middle line:90% Now, here's a fact about Lardil. 00:42:27.270 --> 00:42:30.180 align:middle line:90% It doesn't have monosyllables. 00:42:30.180 --> 00:42:35.360 align:middle line:84% All of its words are at least two syllables long. 00:42:35.360 --> 00:42:37.590 align:middle line:84% So apparently, I think Vlada is absolutely right. 00:42:37.590 --> 00:42:40.220 align:middle line:84% Apparently that's being enforced here. 00:42:40.220 --> 00:42:42.563 align:middle line:90% So indeed, these words-- 00:42:42.563 --> 00:42:44.480 align:middle line:84% a whole bunch of people said versions of this. 00:42:44.480 --> 00:42:46.670 align:middle line:84% Let me just channel what you all said-- these words 00:42:46.670 --> 00:42:48.800 align:middle line:84% at the end, "rain," "tip," and "curve," they 00:42:48.800 --> 00:42:53.210 align:middle line:84% have base forms that don't appear anywhere, in a sense. 00:42:53.210 --> 00:42:55.560 align:middle line:84% Their base forms are "wun," "bel," and "dal," 00:42:55.560 --> 00:42:57.540 align:middle line:84% and that's what you add the accusative to, 00:42:57.540 --> 00:42:59.180 align:middle line:90% by a totally regular rule. 00:42:59.180 --> 00:43:03.170 align:middle line:84% They end in consonants, and so the accusative is in. 00:43:03.170 --> 00:43:05.030 align:middle line:84% And then we need another rule that 00:43:05.030 --> 00:43:14.760 align:middle line:84% says, "Add -da to monosyllables," 00:43:14.760 --> 00:43:16.290 align:middle line:90% something like that. 00:43:16.290 --> 00:43:21.180 align:middle line:84% So if you're in danger of having a monosyllabic Lardil word, 00:43:21.180 --> 00:43:26.580 align:middle line:90% you add -da to monosyllables. 00:43:26.580 --> 00:43:30.900 align:middle line:84% Lardil doesn't like words that are only one syllable long. 00:43:30.900 --> 00:43:33.030 align:middle line:84% This is what's called a minimal word requirement. 00:43:33.030 --> 00:43:34.990 align:middle line:84% They're cross-linguistically quite common. 00:43:34.990 --> 00:43:37.365 align:middle line:84% There are a lot of languages in the world that don't like 00:43:37.365 --> 00:43:38.926 align:middle line:90% their words to be too short. 00:43:38.926 --> 00:43:40.640 align:middle line:90% OK? 00:43:40.640 --> 00:43:42.560 align:middle line:90% All right, good. 00:43:42.560 --> 00:43:43.700 align:middle line:90% So far, so good. 00:43:43.700 --> 00:43:46.820 align:middle line:90% 00:43:46.820 --> 00:43:49.560 align:middle line:84% Right, so what we want to do is posit underlying forms, 00:43:49.560 --> 00:43:52.490 align:middle line:84% which are often the same as the nominative, but not always. 00:43:52.490 --> 00:43:56.090 align:middle line:84% There also these last three that have different forms, OK? 00:43:56.090 --> 00:43:56.820 align:middle line:90% Yes? 00:43:56.820 --> 00:43:59.360 align:middle line:84% AUDIENCE: Then, just to clarify, "wun," "dal," 00:43:59.360 --> 00:44:02.240 align:middle line:90% and "wunda" are all morphemes? 00:44:02.240 --> 00:44:03.470 align:middle line:90% NORVIN RICHARDS: Well, OK. 00:44:03.470 --> 00:44:06.440 align:middle line:90% "Wun" is a morpheme. 00:44:06.440 --> 00:44:11.170 align:middle line:84% "Wunda" up there, stingray species, is a morpheme. 00:44:11.170 --> 00:44:13.660 align:middle line:84% I don't know if we want to call this a morpheme or not. 00:44:13.660 --> 00:44:16.480 align:middle line:84% This is why I was reacting the way I was before. 00:44:16.480 --> 00:44:19.090 align:middle line:90% It doesn't mean anything, right? 00:44:19.090 --> 00:44:23.320 align:middle line:84% It's just there so the word can be long enough. 00:44:23.320 --> 00:44:25.150 align:middle line:90% Yeah, that's all it's for. 00:44:25.150 --> 00:44:29.740 align:middle line:84% So I think linguists would mostly not call it a morpheme. 00:44:29.740 --> 00:44:34.570 align:middle line:84% It's kind of like we said for Polish, 00:44:34.570 --> 00:44:38.140 align:middle line:84% if a word ends with a "g," the "g' becomes a "k." 00:44:38.140 --> 00:44:41.230 align:middle line:84% The change from a "g" to a "k" isn't a morpheme. 00:44:41.230 --> 00:44:44.750 align:middle line:84% It's part of something that Polish prefers about its words. 00:44:44.750 --> 00:44:46.450 align:middle line:84% It doesn't like them to end with "k." 00:44:46.450 --> 00:44:48.730 align:middle line:84% Lardil doesn't like words to be monosyllables, 00:44:48.730 --> 00:44:50.200 align:middle line:90% and this is its repair. 00:44:50.200 --> 00:44:53.570 align:middle line:90% It's how it fixes that problem. 00:44:53.570 --> 00:44:54.850 align:middle line:90% Yeah? 00:44:54.850 --> 00:44:57.410 align:middle line:90% AUDIENCE: I'm confused a little. 00:44:57.410 --> 00:44:58.870 align:middle line:84% NORVIN RICHARDS: I don't blame you. 00:44:58.870 --> 00:45:03.190 align:middle line:84% AUDIENCE: They don't want monosyllables as words, 00:45:03.190 --> 00:45:07.990 align:middle line:84% but they create monosyllable morphemes 00:45:07.990 --> 00:45:13.005 align:middle line:84% and build whole verbs on those two [INAUDIBLE].. 00:45:13.005 --> 00:45:15.130 align:middle line:84% NORVIN RICHARDS: Yeah, I totally see what you mean. 00:45:15.130 --> 00:45:16.990 align:middle line:84% So let me just repeat what you just said. 00:45:16.990 --> 00:45:19.660 align:middle line:84% These people don't like monosyllables, right? 00:45:19.660 --> 00:45:24.910 align:middle line:84% And yet their lexicon is full of monosyllables, right? 00:45:24.910 --> 00:45:27.250 align:middle line:84% It's as though, I don't know, somebody else 00:45:27.250 --> 00:45:29.325 align:middle line:90% gave them their lexicon, right? 00:45:29.325 --> 00:45:30.700 align:middle line:84% And then they're doing their best 00:45:30.700 --> 00:45:33.310 align:middle line:84% to improve it in the ways that will 00:45:33.310 --> 00:45:34.810 align:middle line:90% make them comfortable with it. 00:45:34.810 --> 00:45:37.870 align:middle line:84% Or another way to say it may be when 00:45:37.870 --> 00:45:40.090 align:middle line:84% we say they don't like monosyllables, 00:45:40.090 --> 00:45:44.050 align:middle line:84% they don't like pronouncing monosyllables. 00:45:44.050 --> 00:45:46.845 align:middle line:84% They don't mind monosyllables in their lexicon, 00:45:46.845 --> 00:45:48.970 align:middle line:84% but they're not willing to pronounce them that way. 00:45:48.970 --> 00:45:51.940 align:middle line:84% They have to do something to them, kind of like Poles 00:45:51.940 --> 00:45:54.220 align:middle line:84% don't mind words that end with G as long as they're 00:45:54.220 --> 00:45:55.750 align:middle line:90% in their brains, right? 00:45:55.750 --> 00:45:56.725 align:middle line:90% It's just saying them. 00:45:56.725 --> 00:45:57.850 align:middle line:90% They don't like doing that. 00:45:57.850 --> 00:46:01.570 align:middle line:84% They have to do something to them first. 00:46:01.570 --> 00:46:02.680 align:middle line:90% Nice point, actually. 00:46:02.680 --> 00:46:05.560 align:middle line:84% This is one of many places in linguistics 00:46:05.560 --> 00:46:08.140 align:middle line:84% where you get the feeling that languages are designed 00:46:08.140 --> 00:46:11.533 align:middle line:84% by committees which have different priorities, 00:46:11.533 --> 00:46:14.200 align:middle line:84% that there's one group of people who are like making the lexicon 00:46:14.200 --> 00:46:16.780 align:middle line:84% and another group of people who are like-- 00:46:16.780 --> 00:46:19.202 align:middle line:84% I imagine them taking reports from the first committee. 00:46:19.202 --> 00:46:20.410 align:middle line:90% And then they're like, "What? 00:46:20.410 --> 00:46:21.220 align:middle line:90% You gave us what? 00:46:21.220 --> 00:46:22.810 align:middle line:84% Look at all these monosyllables." 00:46:22.810 --> 00:46:25.680 align:middle line:90% 00:46:25.680 --> 00:46:26.620 align:middle line:90% Yeah? 00:46:26.620 --> 00:46:28.990 align:middle line:90% OK. 00:46:28.990 --> 00:46:32.260 align:middle line:84% So going further into Lardil then. 00:46:32.260 --> 00:46:34.390 align:middle line:90% Here are some more words. 00:46:34.390 --> 00:46:36.220 align:middle line:84% The word for "fish" is "yaka," the word 00:46:36.220 --> 00:46:39.850 align:middle line:84% for "string" is "birrka," and the word for "head" is "lelka." 00:46:39.850 --> 00:46:45.250 align:middle line:84% And the accusatives are "yakin," "birrkin," and "lelkin." 00:46:45.250 --> 00:46:48.090 align:middle line:90% 00:46:48.090 --> 00:46:49.860 align:middle line:84% Is everyone appropriately upset by this? 00:46:49.860 --> 00:46:54.930 align:middle line:84% What would we have expected the accusative of "yaka" to be, 00:46:54.930 --> 00:46:56.320 align:middle line:90% the word for fish? 00:46:56.320 --> 00:46:57.070 align:middle line:90% AUDIENCE: "Yakan?" 00:46:57.070 --> 00:46:58.362 align:middle line:90% NORVIN RICHARDS: "Yakan," yeah. 00:46:58.362 --> 00:47:01.040 align:middle line:90% 00:47:01.040 --> 00:47:03.200 align:middle line:90% What should we do? 00:47:03.200 --> 00:47:03.897 align:middle line:90% Yeah, Joseph? 00:47:03.897 --> 00:47:05.480 align:middle line:84% AUDIENCE: Is there a set of consonants 00:47:05.480 --> 00:47:07.355 align:middle line:84% that you're not allowed to end the word with? 00:47:07.355 --> 00:47:11.680 align:middle line:84% NORVIN RICHARDS: Ah, so why do you ask? 00:47:11.680 --> 00:47:13.680 align:middle line:84% AUDIENCE: Well, they all end with an "a." 00:47:13.680 --> 00:47:15.597 align:middle line:84% NORVIN RICHARDS: So I mean, they don't, right? 00:47:15.597 --> 00:47:19.370 align:middle line:84% "Yaka" ends in a vowel and "yakin" ends in an "n," right? 00:47:19.370 --> 00:47:21.050 align:middle line:90% Yeah. 00:47:21.050 --> 00:47:22.985 align:middle line:84% AUDIENCE: Supposing that they have-- 00:47:22.985 --> 00:47:25.820 align:middle line:84% that the underlying form of "yaka" is "yak," 00:47:25.820 --> 00:47:28.970 align:middle line:84% but you can't end it with "k," so you have to put-- 00:47:28.970 --> 00:47:30.860 align:middle line:84% NORVIN RICHARDS: So these three, if we 00:47:30.860 --> 00:47:32.855 align:middle line:90% went from the accusative-- 00:47:32.855 --> 00:47:33.730 align:middle line:90% what are these three? 00:47:33.730 --> 00:47:35.660 align:middle line:90% "Yaka," "birrka," and "lelka." 00:47:35.660 --> 00:47:38.395 align:middle line:84% If we went from the accusative, if we started-- 00:47:38.395 --> 00:47:40.520 align:middle line:84% if we did the same trick we did with the last three 00:47:40.520 --> 00:47:43.272 align:middle line:84% and said, OK, the nominative, I don't understand why 00:47:43.272 --> 00:47:44.480 align:middle line:90% the nominative is what it is. 00:47:44.480 --> 00:47:46.460 align:middle line:84% But looking just at the accusative, 00:47:46.460 --> 00:47:51.830 align:middle line:84% I wish that the underlying form of "fish" were "yak," right? 00:47:51.830 --> 00:47:52.880 align:middle line:90% That would work. 00:47:52.880 --> 00:47:55.490 align:middle line:84% So our rule for accusatives would get 00:47:55.490 --> 00:47:56.780 align:middle line:90% you the accusatives of "yak." 00:47:56.780 --> 00:47:59.230 align:middle line:84% It would be "yakin," and everything would be perfect. 00:47:59.230 --> 00:48:03.230 align:middle line:84% And then the question is, why isn't the nominative "yak," 00:48:03.230 --> 00:48:05.300 align:middle line:84% to which the answer could be Vlada's answer, 00:48:05.300 --> 00:48:07.580 align:middle line:90% it's only one syllable long. 00:48:07.580 --> 00:48:13.400 align:middle line:84% But then why isn't the nominative of "yak" "yakda?" 00:48:13.400 --> 00:48:15.770 align:middle line:84% And Joseph had a suggestion about that. 00:48:15.770 --> 00:48:20.630 align:middle line:84% "Yak," "birrk," and "lelk" all end in a "k." 00:48:20.630 --> 00:48:23.210 align:middle line:84% So maybe I've got two general rule here. 00:48:23.210 --> 00:48:25.070 align:middle line:84% It's not add -da to monosyllables. 00:48:25.070 --> 00:48:31.700 align:middle line:84% It's add "-da" to, I'll say, most monosyllables. 00:48:31.700 --> 00:48:37.580 align:middle line:90% Add "-a" after "k." 00:48:37.580 --> 00:48:41.120 align:middle line:84% I'll say the same thing over here. 00:48:41.120 --> 00:48:44.330 align:middle line:90% Add -da to most monosyllables. 00:48:44.330 --> 00:48:49.630 align:middle line:90% Add "-a" after "k." 00:48:49.630 --> 00:48:50.130 align:middle line:90% OK? 00:48:50.130 --> 00:48:50.700 align:middle line:90% Yeah, Joseph? 00:48:50.700 --> 00:48:52.617 align:middle line:84% AUDIENCE: Can you not have two stops in a row? 00:48:52.617 --> 00:48:54.943 align:middle line:90% Variable. 00:48:54.943 --> 00:48:57.110 align:middle line:84% NORVIN RICHARDS: You cannot have two stops in a row. 00:48:57.110 --> 00:48:58.190 align:middle line:84% AUDIENCE: You have to have a vowel? 00:48:58.190 --> 00:48:59.240 align:middle line:84% NORVIN RICHARDS: Yup, that would break that. 00:48:59.240 --> 00:49:00.180 align:middle line:90% That's right. 00:49:00.180 --> 00:49:04.490 align:middle line:84% There aren't any Lardil words that have two stops in a row. 00:49:04.490 --> 00:49:06.515 align:middle line:90% Yeah, that's true. 00:49:06.515 --> 00:49:07.015 align:middle line:90% Yeah? 00:49:07.015 --> 00:49:09.980 align:middle line:90% 00:49:09.980 --> 00:49:14.510 align:middle line:84% So yeah, "yak," "birrk," and "lelk," OK? 00:49:14.510 --> 00:49:16.370 align:middle line:84% I think this is the last set of Lardil data 00:49:16.370 --> 00:49:19.460 align:middle line:84% I was going to show you, if I'm remembering right. 00:49:19.460 --> 00:49:20.300 align:middle line:90% No, it's not. 00:49:20.300 --> 00:49:22.530 align:middle line:90% There are some others, sorry. 00:49:22.530 --> 00:49:26.480 align:middle line:84% "Kanda" is blood, "nguka" is water, "ngawa" is dog, 00:49:26.480 --> 00:49:29.300 align:middle line:90% and "karda" is a kinship term. 00:49:29.300 --> 00:49:33.920 align:middle line:84% Lardil kinship terminology is extremely different from ours 00:49:33.920 --> 00:49:36.200 align:middle line:84% in ways that would take a long time to explain. 00:49:36.200 --> 00:49:38.870 align:middle line:84% But "karda" roughly means a woman's child 00:49:38.870 --> 00:49:40.170 align:middle line:90% or a man's sister's child. 00:49:40.170 --> 00:49:43.330 align:middle line:90% 00:49:43.330 --> 00:49:46.390 align:middle line:84% Yeah, no, I won't try to explain that to you. 00:49:46.390 --> 00:49:49.150 align:middle line:84% It's very interesting, but I refuse 00:49:49.150 --> 00:49:51.760 align:middle line:90% to talk about it right now. 00:49:51.760 --> 00:49:56.180 align:middle line:84% So the accusative of "kanda" is "kandun," 00:49:56.180 --> 00:49:57.590 align:middle line:90% which it shouldn't be, right? 00:49:57.590 --> 00:50:01.700 align:middle line:84% It should be "kandan," if it were going to be anything. 00:50:01.700 --> 00:50:02.900 align:middle line:90% Do we want the underlying-- 00:50:02.900 --> 00:50:04.963 align:middle line:84% I mean, we have a rule that adds "-da." 00:50:04.963 --> 00:50:06.755 align:middle line:84% Do we want the underlying form to be "kan?" 00:50:06.755 --> 00:50:12.700 align:middle line:90% 00:50:12.700 --> 00:50:14.170 align:middle line:90% No, right? 00:50:14.170 --> 00:50:15.760 align:middle line:84% If the underlying form were "kan," 00:50:15.760 --> 00:50:18.050 align:middle line:90% what would the accusative be? 00:50:18.050 --> 00:50:18.960 align:middle line:90% AUDIENCE: "Kanin." 00:50:18.960 --> 00:50:20.835 align:middle line:84% NORVIN RICHARDS: "Kanin," yeah, and it's not. 00:50:20.835 --> 00:50:22.200 align:middle line:90% It's "kandun." 00:50:22.200 --> 00:50:23.825 align:middle line:90% So we need another rule. 00:50:23.825 --> 00:50:25.200 align:middle line:84% What's the new rule going to say? 00:50:25.200 --> 00:50:29.280 align:middle line:90% 00:50:29.280 --> 00:50:30.300 align:middle line:90% Yes? 00:50:30.300 --> 00:50:31.650 align:middle line:90% AUDIENCE: Just make a guess. 00:50:31.650 --> 00:50:34.110 align:middle line:90% NORVIN RICHARDS: Yeah. 00:50:34.110 --> 00:50:37.572 align:middle line:84% AUDIENCE: You can't end it in "u," maybe. 00:50:37.572 --> 00:50:39.024 align:middle line:84% So you'd have to change everything 00:50:39.024 --> 00:50:43.965 align:middle line:84% that ends in a "u' to an "a" in the nominative. 00:50:43.965 --> 00:50:47.200 align:middle line:90% [INAUDIBLE] 00:50:47.200 --> 00:50:50.763 align:middle line:84% NORVIN RICHARDS: Right, so we're going 00:50:50.763 --> 00:50:52.180 align:middle line:84% to posit underlying forms-- that's 00:50:52.180 --> 00:50:54.640 align:middle line:84% nice-- we're going to posit underlying forms like "kandu," 00:50:54.640 --> 00:50:56.620 align:middle line:90% "nguku," "ngawu," "kardu." 00:50:56.620 --> 00:51:05.630 align:middle line:84% And then we're going to have a rule, change final "u" to "a." 00:51:05.630 --> 00:51:09.070 align:middle line:84% So "nguku," underlying "nguku," gives you 00:51:09.070 --> 00:51:13.900 align:middle line:84% the accusative "nguku," takes the "n," and you get "ngukun." 00:51:13.900 --> 00:51:15.490 align:middle line:84% But there's another rule that says, 00:51:15.490 --> 00:51:22.130 align:middle line:84% change final "u" to "a," that in the nominative changes 00:51:22.130 --> 00:51:23.260 align:middle line:90% "nguku" to "nguka." 00:51:23.260 --> 00:51:24.190 align:middle line:90% Yeah? 00:51:24.190 --> 00:51:27.020 align:middle line:90% OK, good. 00:51:27.020 --> 00:51:29.980 align:middle line:90% And then-- yup? 00:51:29.980 --> 00:51:30.480 align:middle line:90% All right. 00:51:30.480 --> 00:51:32.490 align:middle line:84% And then I think this is the last set 00:51:32.490 --> 00:51:34.830 align:middle line:90% of data I wanted to show you. 00:51:34.830 --> 00:51:36.660 align:middle line:84% The word for "story" is "ngalu," the word 00:51:36.660 --> 00:51:39.510 align:middle line:84% for "boomerang" is "wangal," the word for "kookaburra" 00:51:39.510 --> 00:51:42.390 align:middle line:84% is "thalkurr," and the word for "umbilical cord" is "kundul," 00:51:42.390 --> 00:51:44.160 align:middle line:84% but the accusatives are "ngalukin," 00:51:44.160 --> 00:51:46.950 align:middle line:84% "wangalkin," "thalkurrkin," and "kundulkin." 00:51:46.950 --> 00:51:50.240 align:middle line:84% What's the underlying form of these? 00:51:50.240 --> 00:51:51.980 align:middle line:84% You guys all know the drill by now. 00:51:51.980 --> 00:51:57.340 align:middle line:84% It's like, take the accusative, subtract the accusative suffix. 00:51:57.340 --> 00:52:02.050 align:middle line:84% So what should the accusative of "umbilical cord" be? 00:52:02.050 --> 00:52:07.290 align:middle line:84% Sorry, what should the bare form of "umbilical cord" be? 00:52:07.290 --> 00:52:09.120 align:middle line:90% "Kundulk," yeah? 00:52:09.120 --> 00:52:10.020 align:middle line:90% "Kundulk." 00:52:10.020 --> 00:52:12.600 align:middle line:84% So we would expect "kundulk," "thalkurrk," "wangalk," 00:52:12.600 --> 00:52:14.310 align:middle line:90% and "ngaluk." 00:52:14.310 --> 00:52:22.540 align:middle line:84% And there's something that says, drop final "k." 00:52:22.540 --> 00:52:27.310 align:middle line:84% So "kundulk," we start with "kundulk," 00:52:27.310 --> 00:52:30.120 align:middle line:84% and we form the accusative off of that. 00:52:30.120 --> 00:52:34.290 align:middle line:84% But if it's nominative, we get rid of the final "k." 00:52:34.290 --> 00:52:35.425 align:middle line:90% Yes? 00:52:35.425 --> 00:52:37.612 align:middle line:84% AUDIENCE: What happened to "birrka?" 00:52:37.612 --> 00:52:39.070 align:middle line:84% NORVIN RICHARDS: What happens what? 00:52:39.070 --> 00:52:40.912 align:middle line:84% AUDIENCE: What happened to "birrka?" 00:52:40.912 --> 00:52:42.370 align:middle line:84% NORVIN RICHARDS: I missed it again. 00:52:42.370 --> 00:52:45.250 align:middle line:84% AUDIENCE: Why is that not "birrka?" 00:52:45.250 --> 00:52:46.810 align:middle line:90% NORVIN RICHARDS: Why is-- 00:52:46.810 --> 00:52:48.190 align:middle line:90% AUDIENCE: I think it's one way. 00:52:48.190 --> 00:52:50.800 align:middle line:84% NORVIN RICHARDS: Oh, the word for string or whatever it was, 00:52:50.800 --> 00:52:54.340 align:middle line:84% "birrka," which started off as "birrk." 00:52:54.340 --> 00:52:56.600 align:middle line:84% And we asked, why aren't we dropping the final K? 00:52:56.600 --> 00:52:58.600 align:middle line:84% Yeah, you're asking exactly the right question-- 00:52:58.600 --> 00:53:00.910 align:middle line:90% or similarly, "ngalu." 00:53:00.910 --> 00:53:02.680 align:middle line:84% We just came up with a rule that changes 00:53:02.680 --> 00:53:04.810 align:middle line:90% final "u" to "a," right? 00:53:04.810 --> 00:53:07.540 align:middle line:84% But the word for story is not "ngala." 00:53:07.540 --> 00:53:09.250 align:middle line:90% It's "ngalu." 00:53:09.250 --> 00:53:10.062 align:middle line:90% Yeah? 00:53:10.062 --> 00:53:12.520 align:middle line:84% AUDIENCE: Do we have to specify that dropping the final "k' 00:53:12.520 --> 00:53:14.685 align:middle line:84% applies to underlying forms internal to swap it? 00:53:14.685 --> 00:53:16.810 align:middle line:84% NORVIN RICHARDS: So we might have to say something. 00:53:16.810 --> 00:53:18.610 align:middle line:84% We either have to work on our rules-- 00:53:18.610 --> 00:53:20.102 align:middle line:84% you're absolutely right-- and make 00:53:20.102 --> 00:53:21.310 align:middle line:90% them more specific than that. 00:53:21.310 --> 00:53:22.540 align:middle line:84% That's one kind of thing we could do. 00:53:22.540 --> 00:53:23.290 align:middle line:90% Yes? 00:53:23.290 --> 00:53:24.100 align:middle line:84% AUDIENCE: And I'm wondering if there's 00:53:24.100 --> 00:53:25.780 align:middle line:84% some kind of like declension system. 00:53:25.780 --> 00:53:27.863 align:middle line:84% Like, I'm thinking if like I had to study Lardil-- 00:53:27.863 --> 00:53:29.155 align:middle line:90% some kind of declension system. 00:53:29.155 --> 00:53:31.280 align:middle line:84% NORVIN RICHARDS: So we couldn't do that for Lardil. 00:53:31.280 --> 00:53:33.290 align:middle line:84% We could say, you know what, there 00:53:33.290 --> 00:53:35.990 align:middle line:84% are nouns that do one of these, and then 00:53:35.990 --> 00:53:38.310 align:middle line:84% there are nouns that do another set of these. 00:53:38.310 --> 00:53:40.970 align:middle line:84% So there are rules that apply to particular classes of nouns. 00:53:40.970 --> 00:53:44.430 align:middle line:84% We could absolutely say that, yeah. 00:53:44.430 --> 00:53:46.140 align:middle line:84% Here's another thing we could do, though. 00:53:46.140 --> 00:53:49.890 align:middle line:84% So here's-- just to stick to the case that you were worried 00:53:49.890 --> 00:53:52.713 align:middle line:90% about just a second ago-- 00:53:52.713 --> 00:53:53.880 align:middle line:90% What did I do with my chalk? 00:53:53.880 --> 00:53:56.490 align:middle line:90% I guess I took it over there. 00:53:56.490 --> 00:54:00.720 align:middle line:84% We've got words like "birrk," that means 00:54:00.720 --> 00:54:03.660 align:middle line:90% string, underlying "birrk." 00:54:03.660 --> 00:54:06.110 align:middle line:84% And then we've got words like underlying "kundulk." 00:54:06.110 --> 00:54:09.280 align:middle line:90% 00:54:09.280 --> 00:54:12.010 align:middle line:84% That means umbilical cord, right? 00:54:12.010 --> 00:54:17.230 align:middle line:84% And with "birrk," what we do is add "a" because-- 00:54:17.230 --> 00:54:18.790 align:middle line:90% so what we do is add "a." 00:54:18.790 --> 00:54:23.530 align:middle line:84% But with "kundulk," what we do is get rid of the final "k." 00:54:23.530 --> 00:54:26.020 align:middle line:84% So these two words are undergoing 00:54:26.020 --> 00:54:28.360 align:middle line:84% different types of rules, different sets of rules. 00:54:28.360 --> 00:54:34.150 align:middle line:84% Or similarly, we had a word, "kandun." 00:54:34.150 --> 00:54:36.220 align:middle line:90% Here, let's do blood, "kandun." 00:54:36.220 --> 00:54:38.360 align:middle line:84% And that, the "u," changed to "a." 00:54:38.360 --> 00:54:40.060 align:middle line:90% So we've got "kanda." 00:54:40.060 --> 00:54:44.320 align:middle line:84% But now we've got a word which is underlying "ngaluk." 00:54:44.320 --> 00:54:46.150 align:middle line:84% And what's happening there is that you're 00:54:46.150 --> 00:54:47.560 align:middle line:90% dropping the final "k." 00:54:47.560 --> 00:54:48.850 align:middle line:90% So you've got "ngalu." 00:54:48.850 --> 00:54:52.420 align:middle line:84% But you don't change the final "u" to an "a." 00:54:52.420 --> 00:54:54.400 align:middle line:84% So we have these rules that seem to apply 00:54:54.400 --> 00:54:56.560 align:middle line:90% to certain bodies of examples. 00:54:56.560 --> 00:54:58.832 align:middle line:84% And your suggestion, which seems attractive, 00:54:58.832 --> 00:55:00.790 align:middle line:84% is maybe these rules need a little bit of work. 00:55:00.790 --> 00:55:03.250 align:middle line:84% Maybe we need to get them to be more specific so they only 00:55:03.250 --> 00:55:06.458 align:middle line:84% apply to certain kinds of things. 00:55:06.458 --> 00:55:08.250 align:middle line:84% But here's another way to think about this, 00:55:08.250 --> 00:55:09.630 align:middle line:90% and I'll leave you with this. 00:55:09.630 --> 00:55:12.450 align:middle line:90% 00:55:12.450 --> 00:55:16.340 align:middle line:84% So here are the rules, some of the rules. 00:55:16.340 --> 00:55:19.090 align:middle line:84% So yeah, the word for "head" starts off as "lelk." 00:55:19.090 --> 00:55:23.370 align:middle line:84% And we add "a" because of rule two, so we get "lelka." 00:55:23.370 --> 00:55:28.770 align:middle line:84% But this is what we're asking, the kind of thing we're asking. 00:55:28.770 --> 00:55:30.030 align:middle line:90% You start with "lelk." 00:55:30.030 --> 00:55:33.240 align:middle line:84% Why don't you apply the rule four to get "lel," 00:55:33.240 --> 00:55:36.510 align:middle line:84% and then maybe rule one, so you'd get "lelda?" 00:55:36.510 --> 00:55:38.010 align:middle line:90% Or you start with "ngaluk." 00:55:38.010 --> 00:55:40.710 align:middle line:84% You apply rule four, so you're getting "ngalu." 00:55:40.710 --> 00:55:44.340 align:middle line:84% Why don't you then apply rule three to get "ngala?" 00:55:44.340 --> 00:55:46.510 align:middle line:84% These are the kinds of questions that we're asking. 00:55:46.510 --> 00:55:49.080 align:middle line:84% And one kind of response to this kind of problem that people 00:55:49.080 --> 00:55:52.350 align:middle line:84% have classically given is to not only have rules 00:55:52.350 --> 00:55:55.320 align:middle line:84% but to have the rules apply in an order, that is, 00:55:55.320 --> 00:55:56.940 align:middle line:84% to have what's called rule ordering. 00:55:56.940 --> 00:55:59.460 align:middle line:84% It's as though there's an assembly line between one 00:55:59.460 --> 00:56:01.530 align:middle line:84% of these committees and the other. 00:56:01.530 --> 00:56:08.340 align:middle line:84% So rule two and rule three happened before rule four. 00:56:08.340 --> 00:56:13.470 align:middle line:84% So if you are starting with "lelk," 00:56:13.470 --> 00:56:16.110 align:middle line:84% the basic form for "head," you first 00:56:16.110 --> 00:56:19.830 align:middle line:90% apply rule two and get "lelka." 00:56:19.830 --> 00:56:23.520 align:middle line:84% Having applied rule two, which adds the "a' after "lelk," 00:56:23.520 --> 00:56:25.320 align:middle line:84% you're not in any danger of applying 00:56:25.320 --> 00:56:26.610 align:middle line:90% rule three or rule four. 00:56:26.610 --> 00:56:28.658 align:middle line:84% And rule three, never mind, but rule four, 00:56:28.658 --> 00:56:30.450 align:middle line:84% which would have gotten rid of a final "k," 00:56:30.450 --> 00:56:32.700 align:middle line:84% doesn't apply anymore because, well, there 00:56:32.700 --> 00:56:34.890 align:middle line:90% isn't a final "k" anymore. 00:56:34.890 --> 00:56:40.470 align:middle line:84% Or "ngaluk," which is the base form for story, rule four 00:56:40.470 --> 00:56:43.680 align:middle line:84% happens after rule two and rule three. 00:56:43.680 --> 00:56:48.120 align:middle line:84% So rule four gets rid of the "k' and gives you "ngalu." 00:56:48.120 --> 00:56:50.370 align:middle line:84% And you can imagine now at this point, 00:56:50.370 --> 00:56:52.710 align:middle line:84% the rule three people are like, hey, give us that word. 00:56:52.710 --> 00:56:53.760 align:middle line:90% We want to get rid of that "u." 00:56:53.760 --> 00:56:54.570 align:middle line:90% But it's too late. 00:56:54.570 --> 00:56:57.540 align:middle line:90% Rule three has had its time. 00:56:57.540 --> 00:56:59.050 align:middle line:84% You don't get to apply it anymore. 00:56:59.050 --> 00:57:01.620 align:middle line:84% So this is one way of handling these kinds of problems, 00:57:01.620 --> 00:57:04.130 align:middle line:84% is to see, these rules that change things, 00:57:04.130 --> 00:57:05.880 align:middle line:84% they're not just rules that change things. 00:57:05.880 --> 00:57:07.590 align:middle line:84% They're rules that apply in an order. 00:57:07.590 --> 00:57:10.830 align:middle line:84% There's a series of processes that you apply. 00:57:10.830 --> 00:57:15.090 align:middle line:84% And you have to apply them in the correct order, 00:57:15.090 --> 00:57:18.030 align:middle line:84% and not in the incorrect order, to get the result. Joseph? 00:57:18.030 --> 00:57:20.970 align:middle line:84% AUDIENCE: So it's better to do these like a series of rules 00:57:20.970 --> 00:57:22.640 align:middle line:84% that are all applied simultaneously, 00:57:22.640 --> 00:57:24.660 align:middle line:84% in a sequential if/else statement. 00:57:24.660 --> 00:57:24.930 align:middle line:90% NORVIN RICHARDS: Right. 00:57:24.930 --> 00:57:27.090 align:middle line:84% AUDIENCE: And when you're done, you need to [INAUDIBLE].. 00:57:27.090 --> 00:57:29.215 align:middle line:84% NORVIN RICHARDS: That is a way to think about this, 00:57:29.215 --> 00:57:30.007 align:middle line:90% yes, a classic way. 00:57:30.007 --> 00:57:31.507 align:middle line:84% We'll talk about other ways to think 00:57:31.507 --> 00:57:33.341 align:middle line:84% about this kind of problem, but this is one. 00:57:33.341 --> 00:57:35.570 align:middle line:90% All right, let's stop there. 00:57:35.570 --> 00:57:39.000 align:middle line:90%