1 00:00:00,430 --> 00:00:04,060 The following content is provided by MIT OpenCourseWare 2 00:00:04,060 --> 00:00:06,160 under a Creative Commons license. 3 00:00:06,160 --> 00:00:09,070 Additional information about our license and MIT 4 00:00:09,070 --> 00:00:12,050 OpenCourseWare in general, is available at ocw.mit.edu. 5 00:00:15,160 --> 00:00:18,260 PROFESSOR: In the last lecture I was talking about the fact 6 00:00:18,260 --> 00:00:26,050 that kids go through roughly the stages of language 7 00:00:26,050 --> 00:00:27,380 development. 8 00:00:27,380 --> 00:00:28,920 They go through the same set of 9 00:00:28,920 --> 00:00:31,270 language development stages. 10 00:00:38,090 --> 00:00:42,200 A babbling sort of free verbal stage, one word, two word, and 11 00:00:42,200 --> 00:00:45,940 then more than two word stages. 12 00:00:45,940 --> 00:00:48,610 The question we're left with at the moment is well, what is 13 00:00:48,610 --> 00:00:50,780 it that they're actually learning when 14 00:00:50,780 --> 00:00:52,030 they're learning language. 15 00:00:52,030 --> 00:00:55,430 There's lots of stuff that's got to be learned. 16 00:00:55,430 --> 00:00:58,100 One of the things that, it's on the handout somewhere there 17 00:00:58,100 --> 00:01:00,740 as I recall, is they gotta do a remarkable job of learning 18 00:01:00,740 --> 00:01:02,590 words and word meaning. 19 00:01:02,590 --> 00:01:07,100 Between about a year and a half and six years of age, in 20 00:01:07,100 --> 00:01:15,910 order to account for the vocabulary that a six year old 21 00:01:15,910 --> 00:01:19,870 has, you have to assume that children are learning five to 22 00:01:19,870 --> 00:01:23,190 10 words a day, every day. 23 00:01:23,190 --> 00:01:27,140 So, that's a rate sort of approaching a 24 00:01:27,140 --> 00:01:30,580 word an hour or so. 25 00:01:30,580 --> 00:01:32,110 They're not doing it in school, right. 26 00:01:32,110 --> 00:01:34,320 It's not that they're going to school and every day they're 27 00:01:34,320 --> 00:01:36,160 being handed long vocabulary lists. 28 00:01:36,160 --> 00:01:38,080 They're just picking this stuff up. 29 00:01:38,080 --> 00:01:43,800 It's an amazing and not trivial feat for figuring out 30 00:01:43,800 --> 00:01:45,050 how you do that. 31 00:01:47,900 --> 00:01:52,470 Somebody says oh, look, there's the chalk, and the kid 32 00:01:52,470 --> 00:01:56,470 not only has to learn this new word, chalk, but has to figure 33 00:01:56,470 --> 00:02:04,570 out that it somehow refers to this box of stuff, and not to 34 00:02:04,570 --> 00:02:12,320 the color yellow or to the letter c or to the small boxes 35 00:02:12,320 --> 00:02:15,920 when Bush is president, or any of a vast number of other 36 00:02:15,920 --> 00:02:16,700 possibilities. 37 00:02:16,700 --> 00:02:19,350 It's a very daunting problem. 38 00:02:19,350 --> 00:02:22,460 I'm going to focus rother than on the word meaning problem, 39 00:02:22,460 --> 00:02:25,330 on a different problem, which is the 40 00:02:25,330 --> 00:02:28,660 problem of learning grammar. 41 00:02:31,580 --> 00:02:35,150 There, there are sort of two large scale classes of theory 42 00:02:35,150 --> 00:02:37,280 that I have cartooned -- 43 00:02:37,280 --> 00:02:42,910 I cartooned them on the board and described them on the 44 00:02:42,910 --> 00:02:46,870 handout as the East coast view and the West coast view. 45 00:02:50,060 --> 00:02:52,860 The East coast view could really be called the MIT view 46 00:02:52,860 --> 00:02:53,460 originally. 47 00:02:53,460 --> 00:02:57,270 It really starts with people like Chomsky here and is 48 00:02:57,270 --> 00:03:00,830 nicely wrapped up in Steve Pinker's book called The 49 00:03:00,830 --> 00:03:03,280 Language Instinct, which if you turn out to be interested 50 00:03:03,280 --> 00:03:06,080 in the topic of language and language learning, I'd 51 00:03:06,080 --> 00:03:07,630 strongly recommend you read it. 52 00:03:07,630 --> 00:03:10,290 Among other things, Pinker is one of the better writers in 53 00:03:10,290 --> 00:03:13,700 our field and it's a good read. 54 00:03:16,520 --> 00:03:20,960 The West coast view was a view you held by people like Mark 55 00:03:20,960 --> 00:03:22,470 Seidenberg at UCLA. 56 00:03:22,470 --> 00:03:25,180 Well, Mark Seidenberg's now at Minnesota, so this 57 00:03:25,180 --> 00:03:29,310 geographical thing doesn't work so well anymore. 58 00:03:29,310 --> 00:03:30,560 Here the distinction. 59 00:03:35,100 --> 00:03:36,780 Let's start with the specific problem. 60 00:03:36,780 --> 00:03:41,340 Specific problem might be something like, as it says on 61 00:03:41,340 --> 00:03:49,370 the handout, how do you make the past tense in English? 62 00:03:49,370 --> 00:03:51,350 You'd obviously have to learn to make the past tense in some 63 00:03:51,350 --> 00:03:54,220 other language if you were going to learn yourself some 64 00:03:54,220 --> 00:03:56,420 other language. 65 00:03:56,420 --> 00:03:59,470 But I walked to the store. 66 00:03:59,470 --> 00:04:01,100 Yesterday I did what? 67 00:04:01,100 --> 00:04:03,070 AUDIENCE: [INAUDIBLE]. 68 00:04:03,070 --> 00:04:07,040 PROFESSOR: I walked to the store. 69 00:04:07,040 --> 00:04:14,590 Today I plan to go to the airport, actually, right after 70 00:04:14,590 --> 00:04:16,050 this lecture. 71 00:04:16,050 --> 00:04:17,480 Yesterday I did what? 72 00:04:17,480 --> 00:04:18,650 AUDIENCE: [INAUDIBLE]. 73 00:04:18,650 --> 00:04:20,220 PROFESSOR: All right. 74 00:04:20,220 --> 00:04:24,110 So, how do we go and make the past tense in English? 75 00:04:24,110 --> 00:04:28,530 AUDIENCE: e-d. 76 00:04:28,530 --> 00:04:31,780 PROFESSOR: Yeah, stick an e-d on things most of the time, 77 00:04:31,780 --> 00:04:37,070 except today I'm going to sing yesterday I--. 78 00:04:37,070 --> 00:04:37,250 [UNINTELLIGIBLE] 79 00:04:37,250 --> 00:04:40,030 I didn't singed. 80 00:04:40,030 --> 00:04:42,510 Today we ring the bell, yesterday-- 81 00:04:42,510 --> 00:04:43,270 AUDIENCE: Rang. 82 00:04:43,270 --> 00:04:45,200 PROFESSOR: We rang the bell, or we rung the but -- no we 83 00:04:45,200 --> 00:04:47,730 didn't rung the bell yesterday. 84 00:04:47,730 --> 00:04:50,250 The bell was rung yesterday, but we rang it--. 85 00:04:50,250 --> 00:04:53,400 Anyway, it gets complicated. 86 00:04:53,400 --> 00:04:59,570 Today I bring you all this information about language, 87 00:04:59,570 --> 00:05:00,820 and yesterday I--. 88 00:05:00,820 --> 00:05:01,660 AUDIENCE: Brought. 89 00:05:01,660 --> 00:05:04,180 PROFESSOR: Brought -- it's not even brang. 90 00:05:04,180 --> 00:05:07,890 So the rule -- there's sort of a sub sing-sang kind of rule, 91 00:05:07,890 --> 00:05:10,140 there's a bring-brang. 92 00:05:10,140 --> 00:05:11,040 No, brang doesn't work. 93 00:05:11,040 --> 00:05:13,470 Brought. 94 00:05:13,470 --> 00:05:20,660 So, kids behave as thought they have learned a rule and a 95 00:05:20,660 --> 00:05:22,190 set of exceptions. 96 00:05:22,190 --> 00:05:26,110 The East coast view of this is that they really 97 00:05:26,110 --> 00:05:28,470 have learned a rule. 98 00:05:28,470 --> 00:05:34,160 That somewhere deep in their little brain, you could 99 00:05:34,160 --> 00:05:38,300 hypothetically go in and find the rule, that it 100 00:05:38,300 --> 00:05:42,260 is written in there. 101 00:05:42,260 --> 00:05:47,280 One of the lines of evidence for this is that at a certain 102 00:05:47,280 --> 00:05:50,410 point during development as they're learning this, they 103 00:05:50,410 --> 00:05:53,900 start to overuse the rule. 104 00:05:53,900 --> 00:06:03,770 A child will say things like I goed to the store yesterday, 105 00:06:03,770 --> 00:06:09,370 or I bringed the ball home from school yesterday, or 106 00:06:09,370 --> 00:06:10,070 something like that. 107 00:06:10,070 --> 00:06:11,780 They'll start to overgeneralize. 108 00:06:11,780 --> 00:06:13,170 They don't do it all the time. 109 00:06:13,170 --> 00:06:17,940 But they start to behave as though they are saying ah-ha, 110 00:06:17,940 --> 00:06:19,770 got this verb here, I know I need to make it 111 00:06:19,770 --> 00:06:20,590 into the past tense. 112 00:06:20,590 --> 00:06:23,210 What's the rule for making the past tense? 113 00:06:23,210 --> 00:06:26,290 I gotta stick an e-d on the end of it. 114 00:06:26,290 --> 00:06:28,540 Even though a few months earlier they might have 115 00:06:28,540 --> 00:06:31,870 successfully said went. 116 00:06:31,870 --> 00:06:35,150 It is as if they learn a rule, and they learn a set of 117 00:06:35,150 --> 00:06:38,090 exceptions. 118 00:06:38,090 --> 00:06:41,095 When you go in and look for what you're trying to say, 119 00:06:41,095 --> 00:06:43,350 past tense of go. 120 00:06:43,350 --> 00:06:47,630 If the past tense of go, you know, I went, doesn't get 121 00:06:47,630 --> 00:06:50,120 there fast enough, you default to the rule. 122 00:06:50,120 --> 00:06:54,800 You say I goed. 123 00:06:54,800 --> 00:07:02,070 This assumes a certain amount of stuff in the head ahead of 124 00:07:02,070 --> 00:07:03,760 time, that's what this cartoon is about. 125 00:07:03,760 --> 00:07:09,350 It doesn't need to be exactly this kind of stuff, but this 126 00:07:09,350 --> 00:07:14,020 is the sort of stuff that you might have in the head. 127 00:07:14,020 --> 00:07:17,090 Let's think, actually, going back to the lectures I was 128 00:07:17,090 --> 00:07:20,120 giving on cognitive development last week. 129 00:07:20,120 --> 00:07:23,440 Remember, there's pretty good evidence that little kids have 130 00:07:23,440 --> 00:07:26,080 in their head an idea about objects. 131 00:07:26,080 --> 00:07:28,250 They know there are such things about objects. 132 00:07:28,250 --> 00:07:31,800 So -- oops, I didn't put objects in here. 133 00:07:31,800 --> 00:07:36,180 Let's assume that there are such things as objects. 134 00:07:36,180 --> 00:07:41,040 Also, we saw that little kids know something about number, 135 00:07:41,040 --> 00:07:43,630 like they can do that Karen Wynn style baby math. 136 00:07:43,630 --> 00:07:48,080 They know that there's 1 and more than 1, certainly. 137 00:07:48,080 --> 00:07:52,110 Now let's assume that they've got some innate idea, some 138 00:07:52,110 --> 00:07:54,120 innate capability to learn language. 139 00:07:54,120 --> 00:07:56,410 Well, if I'm going to learn a language of any use, it's 140 00:07:56,410 --> 00:07:57,680 going to have to be able to talk about 141 00:07:57,680 --> 00:07:59,760 these object things. 142 00:07:59,760 --> 00:08:03,170 So I'm going to have some sort of innate idea that 143 00:08:03,170 --> 00:08:04,900 there will be nouns. 144 00:08:04,900 --> 00:08:07,430 I don't know what the nouns are going to look like in this 145 00:08:07,430 --> 00:08:12,520 language, but I come into the world in this view knowing 146 00:08:12,520 --> 00:08:15,200 that there have to be noun-like things. 147 00:08:15,200 --> 00:08:20,570 Not only that, I know that there can be one thing and 148 00:08:20,570 --> 00:08:22,200 there can be more than one thing. 149 00:08:22,200 --> 00:08:25,200 It therefore follows that I need to be able to 150 00:08:25,200 --> 00:08:26,930 talk about that, too. 151 00:08:26,930 --> 00:08:32,700 And so my job as a little language learner is to figure 152 00:08:32,700 --> 00:08:37,830 out what do the nouns look like, and how do I make those 153 00:08:37,830 --> 00:08:43,260 into plural in my language. 154 00:08:43,260 --> 00:08:45,820 I said there's an infinite number of ways in principle 155 00:08:45,820 --> 00:08:47,790 that I could designate a plural. 156 00:08:47,790 --> 00:08:52,150 But I will learn that, for the most part, in English it's 157 00:08:52,150 --> 00:08:56,320 going to be add s, right? 158 00:08:56,320 --> 00:08:58,780 Oh, look, there's the cat. 159 00:08:58,780 --> 00:09:01,970 Oh, look, there are the cats and so on. 160 00:09:01,970 --> 00:09:03,670 I'm going to have to do the same thing as I was talking 161 00:09:03,670 --> 00:09:04,580 about in the past tense. 162 00:09:04,580 --> 00:09:07,150 I'm going to have to learn there are exceptions. 163 00:09:07,150 --> 00:09:12,990 Yeah, cat, cats, rat, rats, chair, chairs, mouse -- 164 00:09:12,990 --> 00:09:14,760 uh, mice. 165 00:09:14,760 --> 00:09:17,960 Fish -- for some reason fish is fish. 166 00:09:17,960 --> 00:09:20,470 One fish, two fish, red fish, blue fish -- all 167 00:09:20,470 --> 00:09:22,370 that kind of stuff. 168 00:09:22,370 --> 00:09:26,250 So again, I'll learn a rule, I'll overgeneralize that rule 169 00:09:26,250 --> 00:09:30,550 when I get to be three or four years old, and I'll learn a 170 00:09:30,550 --> 00:09:33,040 set of exceptions. 171 00:09:33,040 --> 00:09:38,080 I will find when I become a literature professor later on 172 00:09:38,080 --> 00:09:43,010 and study the history of English literature, that 173 00:09:43,010 --> 00:09:45,690 things shift over time -- things that were exceptions or 174 00:09:45,690 --> 00:09:48,600 things that were ungrammatical once upon a time might become 175 00:09:48,600 --> 00:09:49,430 grammatical. 176 00:09:49,430 --> 00:09:52,650 It's a moving target language. 177 00:09:52,650 --> 00:09:56,150 You have to figure out what the current state of your 178 00:09:56,150 --> 00:09:59,430 language might be. 179 00:09:59,430 --> 00:10:04,100 The alternative is the West coast view, which is really 180 00:10:04,100 --> 00:10:07,520 sort of a neural net parallel distributed processing kind of 181 00:10:07,520 --> 00:10:10,700 view is to say that we don't have to propose that there's 182 00:10:10,700 --> 00:10:14,510 all this stuff, these structures in the head in 183 00:10:14,510 --> 00:10:15,260 quite the same way. 184 00:10:15,260 --> 00:10:18,960 What we can say is look, there's this very powerful 185 00:10:18,960 --> 00:10:23,940 association learner in here, a device out there looking for 186 00:10:23,940 --> 00:10:25,670 statistical regularity. 187 00:10:25,670 --> 00:10:27,090 There's evidence -- 188 00:10:27,090 --> 00:10:31,690 I think I mentioned last time that little kids long before 189 00:10:31,690 --> 00:10:35,630 they can speak a language are already sensitive to the 190 00:10:35,630 --> 00:10:38,500 statistical regularities of what syllables can follow what 191 00:10:38,500 --> 00:10:44,910 other syllables in the noises that they are hearing. 192 00:10:44,910 --> 00:10:49,150 So you've got this very powerful statistical learning 193 00:10:49,150 --> 00:10:50,720 devise in your head. 194 00:10:50,720 --> 00:10:55,680 It ends up behaving as if it had a rule, but the rule is 195 00:10:55,680 --> 00:11:01,730 not written out anywhere in brainware. 196 00:11:01,730 --> 00:11:04,900 It's just the way the network behaves. 197 00:11:04,900 --> 00:11:10,875 In the same way that this hunk of chalk -- where's somebody 198 00:11:10,875 --> 00:11:12,430 who looks reasonably alert. 199 00:11:12,430 --> 00:11:14,530 Do you look reasonably alert? 200 00:11:14,530 --> 00:11:15,780 No, now she's looking elsewhere. 201 00:11:18,850 --> 00:11:20,200 All right, how about if I just throw it out. 202 00:11:20,200 --> 00:11:22,870 There we go, look at that. 203 00:11:22,870 --> 00:11:23,510 Very good. 204 00:11:23,510 --> 00:11:24,650 Send it back. 205 00:11:24,650 --> 00:11:25,930 That's a smart hunk of chalk. 206 00:11:25,930 --> 00:11:30,210 It knew to fly on a parabolic path, right, knew a whole 207 00:11:30,210 --> 00:11:32,320 bunch of 801 style physics. 208 00:11:32,320 --> 00:11:34,820 It didn't know it in any real sense, of course. 209 00:11:34,820 --> 00:11:40,040 It behaved as if, because it's operating under simple 210 00:11:40,040 --> 00:11:43,240 physical forces in the world and ends up behaving as if it 211 00:11:43,240 --> 00:11:44,750 knows physics. 212 00:11:44,750 --> 00:11:48,160 It's a somewhat stretched analogy, but the same sort of 213 00:11:48,160 --> 00:11:50,120 thing is going on here. 214 00:11:50,120 --> 00:11:54,520 You learn about the linguistic world you are in and then you 215 00:11:54,520 --> 00:12:00,720 behave as if you knew these rules, rather than having some 216 00:12:00,720 --> 00:12:06,640 sort of mechanism by which you actually, in a sense, at birth 217 00:12:06,640 --> 00:12:09,390 knew that there should be such rules and merely needed to 218 00:12:09,390 --> 00:12:13,110 figure out the particular form that the rule 219 00:12:13,110 --> 00:12:15,040 took in your language. 220 00:12:15,040 --> 00:12:25,290 These are, in a sense, modern and not hugely different views 221 00:12:25,290 --> 00:12:28,780 of initial positions that are radically different. 222 00:12:28,780 --> 00:12:34,070 So, if this is a continuum here, way over here somewhere 223 00:12:34,070 --> 00:12:40,050 would be radical nativist views of language. 224 00:12:40,050 --> 00:12:43,020 Way over here somewhere would be radical 225 00:12:43,020 --> 00:12:47,860 imperacist views of language. 226 00:12:53,800 --> 00:12:57,500 The most radical nativist views nobody particularly 227 00:12:57,500 --> 00:12:59,920 held, except I think I mentioned that little children 228 00:12:59,920 --> 00:13:03,840 are subject to this as an erroneous thought. 229 00:13:03,840 --> 00:13:10,110 That's the view that you are not only born to have 230 00:13:10,110 --> 00:13:13,090 language, but you're born to have a specific language. 231 00:13:13,090 --> 00:13:20,550 So that if you are a Saudi kid, that you are born to 232 00:13:20,550 --> 00:13:24,150 speak Arabic and that's all else being equa -- well, even 233 00:13:24,150 --> 00:13:25,200 all else not being equal. 234 00:13:25,200 --> 00:13:26,280 You're a Saudi kid. 235 00:13:26,280 --> 00:13:30,420 You get transplanted to Sweden, and at the right 236 00:13:30,420 --> 00:13:33,120 moment Arabic appears. 237 00:13:33,120 --> 00:13:33,900 That's not true. 238 00:13:33,900 --> 00:13:37,440 Nobody other than a small kid I think ever particularly 239 00:13:37,440 --> 00:13:39,390 believed this, because it was clearly falsifiable. 240 00:13:43,770 --> 00:13:46,980 The earliest experiment I know in this area, and I don't know 241 00:13:46,980 --> 00:13:49,690 if the experiment was ever done, is something reported by 242 00:13:49,690 --> 00:13:56,000 Herodotus in his book about traveling in Egypt. 243 00:13:56,000 --> 00:14:00,960 He says that King Psammetichos of Egypt 2,500 years ago 244 00:14:00,960 --> 00:14:10,790 believed that there was an original language of humankind 245 00:14:10,790 --> 00:14:16,410 where King Psammetichos had grown up in the Judeo 246 00:14:16,410 --> 00:14:18,770 Christian tradition, he might have called it the language of 247 00:14:18,770 --> 00:14:19,880 the Garden of Eden. 248 00:14:19,880 --> 00:14:21,950 What language did Adam and Eve speak in the garden? 249 00:14:21,950 --> 00:14:26,660 Well, the original language of humankind. 250 00:14:26,660 --> 00:14:31,540 Psammetichos figured he would go and figure out what that 251 00:14:31,540 --> 00:14:32,160 language was. 252 00:14:32,160 --> 00:14:34,180 He actually had a hypothesis about what it was. 253 00:14:34,180 --> 00:14:35,400 He was King of Egypt. 254 00:14:35,400 --> 00:14:39,660 It stood to reason that the original language of humanity 255 00:14:39,660 --> 00:14:41,620 was Egyptian. 256 00:14:41,620 --> 00:14:44,980 So what he is alleged to have done is to take two children, 257 00:14:44,980 --> 00:14:48,160 give them to a shepherd, tell the shepherd to raise them 258 00:14:48,160 --> 00:14:52,890 with sheep and not talk to them, and to report to the 259 00:14:52,890 --> 00:14:56,540 King what the first word was that these children spoke. 260 00:14:59,490 --> 00:15:02,710 The report is that when the children first spoke, their 261 00:15:02,710 --> 00:15:07,720 first word was "beccau," which sounds suspiciously like they 262 00:15:07,720 --> 00:15:09,360 were imitating the sheep I think. 263 00:15:09,360 --> 00:15:14,840 But anyway, beccau turned out to mean bread, and that seemed 264 00:15:14,840 --> 00:15:18,350 like a pretty encouraging thing, except that it meant 265 00:15:18,350 --> 00:15:23,470 bread in Phrygian, the language of Phrygia, which is 266 00:15:23,470 --> 00:15:25,210 in modern day Turkey. 267 00:15:25,210 --> 00:15:29,210 It wasn't word for bread in Egyptian at all. 268 00:15:29,210 --> 00:15:32,020 So, Psammetichos is reputed to have been disappointed by the 269 00:15:32,020 --> 00:15:33,570 results of the experiment. 270 00:15:33,570 --> 00:15:36,370 Look, we don't know that the experiment was ever done. 271 00:15:36,370 --> 00:15:40,040 We do know or we're pretty sure we know how the 272 00:15:40,040 --> 00:15:45,350 experiment would have come out had it actually been done. 273 00:15:45,350 --> 00:15:51,190 Which is that they wouldn't have spoken a language at all. 274 00:15:51,190 --> 00:15:55,940 That in the absence of a linguistic environment, even 275 00:15:55,940 --> 00:15:59,180 if you come into the world with some native endowment to 276 00:15:59,180 --> 00:16:02,640 learn a language, which does seem to be the case, you do 277 00:16:02,640 --> 00:16:07,080 not speak a normal human language unless you are raised 278 00:16:07,080 --> 00:16:11,510 in a linguistic environment at some point. 279 00:16:11,510 --> 00:16:17,370 We think we know this from an isolated set of cases over the 280 00:16:17,370 --> 00:16:23,330 course of historical record where kids are somehow 281 00:16:23,330 --> 00:16:25,850 deprived of linguistic input. 282 00:16:25,850 --> 00:16:31,940 Now that happens typically in odd and disturbing kinds of 283 00:16:31,940 --> 00:16:36,400 circumstances, and so it's nothing like a real experiment 284 00:16:36,400 --> 00:16:38,240 because of course, you'd never do the real experiment, it 285 00:16:38,240 --> 00:16:39,850 wouldn't be even close to ethical. 286 00:16:43,430 --> 00:16:46,620 This class of children are known as wolf children, a term 287 00:16:46,620 --> 00:16:51,540 that has always disturbed me personally. 288 00:16:51,540 --> 00:16:57,910 But the illusion is not to wolfe with an e, but to the 289 00:16:57,910 --> 00:17:01,620 legendary story of the founding of Rome where the 290 00:17:01,620 --> 00:17:05,280 twins Romulus and Remus are abandoned -- 291 00:17:05,280 --> 00:17:07,350 I can't remember why -- but anyway they get 292 00:17:07,350 --> 00:17:09,610 suckled by a she-wolf. 293 00:17:09,610 --> 00:17:12,680 So if you look at Roman coins and things you'll see this 294 00:17:12,680 --> 00:17:17,130 picture of a great big wolf and two little babies suckling 295 00:17:17,130 --> 00:17:19,490 underneath. 296 00:17:19,490 --> 00:17:23,380 Presumably, the wolf wasn't speaking good Latin, but 297 00:17:23,380 --> 00:17:26,350 presumably Romulus and Remus managed to come up with good 298 00:17:26,350 --> 00:17:31,260 Latin somehow out of all this. 299 00:17:33,930 --> 00:17:37,860 Occasionally it happens that some kid gets abandons in the 300 00:17:37,860 --> 00:17:39,400 woods and somehow survives. 301 00:17:39,400 --> 00:17:42,210 I think there's the case of an Indian child who's described 302 00:17:42,210 --> 00:17:43,780 in [? Gleitman ?]. 303 00:17:43,780 --> 00:17:47,620 There was a very interesting but disturbing book called 304 00:17:47,620 --> 00:17:51,780 Genie, written a few years ago, about a child of deeply 305 00:17:51,780 --> 00:17:54,530 disturbed parents who had taken her basically locked her 306 00:17:54,530 --> 00:17:58,540 in a closet for all of her childhood. 307 00:17:58,540 --> 00:18:00,990 The bottom line is that this is not going to be good for 308 00:18:00,990 --> 00:18:03,380 any aspect of your development, but you don't 309 00:18:03,380 --> 00:18:04,930 develop a normal language. 310 00:18:04,930 --> 00:18:08,450 If you don't get exposed to language during a so-called 311 00:18:08,450 --> 00:18:12,610 critical period, which for language is certainly over by 312 00:18:12,610 --> 00:18:17,650 early adolescence and is getting pretty much over, six, 313 00:18:17,650 --> 00:18:19,500 seven, eight years old you'd better get your language 314 00:18:19,500 --> 00:18:24,620 experience in there before that. 315 00:18:24,620 --> 00:18:25,960 You're just never going to speak a 316 00:18:25,960 --> 00:18:27,750 normal grammatic language. 317 00:18:27,750 --> 00:18:29,130 The door closes on it. 318 00:18:29,130 --> 00:18:34,540 So, Genie, this poor girl who was raised in a closet, could 319 00:18:34,540 --> 00:18:38,000 be taught a certain amount of vocabulary and things like 320 00:18:38,000 --> 00:18:45,020 that, but could never somehow get a real grip on the grammar 321 00:18:45,020 --> 00:18:46,150 of a language. 322 00:18:46,150 --> 00:18:48,790 It also turns out to be easier to learn a second language 323 00:18:48,790 --> 00:18:52,510 during the critical period, but not impossible to learn 324 00:18:52,510 --> 00:18:55,850 one afterwards, as those of you have just started learning 325 00:18:55,850 --> 00:18:58,290 French or something will be happy to know. 326 00:18:58,290 --> 00:19:04,030 The claim is that it's very difficult to become a truly 327 00:19:04,030 --> 00:19:14,300 fluent speaker of a second language later in life. 328 00:19:14,300 --> 00:19:16,510 But it's not impossible in the way it appears to be 329 00:19:16,510 --> 00:19:22,380 impossible to learn a first language after the end of the 330 00:19:22,380 --> 00:19:25,220 critical period. 331 00:19:25,220 --> 00:19:34,040 The alternative, the radical alternative, to King 332 00:19:34,040 --> 00:19:36,850 Psammetichos on one side, if we're sticking with classical 333 00:19:36,850 --> 00:19:38,540 references. 334 00:19:38,540 --> 00:19:41,110 On this side would be somebody like Saint Augustine who 335 00:19:41,110 --> 00:19:43,890 writes that he knows how he learned language. 336 00:19:43,890 --> 00:19:49,650 He learned language by being explicitly taught. 337 00:19:49,650 --> 00:19:51,470 The same way you'd learn anything else. 338 00:19:51,470 --> 00:20:00,490 That people said that's chalk, that's a chair, that's your TA 339 00:20:00,490 --> 00:20:03,690 and you learned this sort of stuff. 340 00:20:03,690 --> 00:20:06,950 That it was an act of learning not unlike 341 00:20:06,950 --> 00:20:08,280 other acts of learning. 342 00:20:08,280 --> 00:20:10,420 This gets developed -- 343 00:20:10,420 --> 00:20:14,310 Saint Augustine was not particularly working on 344 00:20:14,310 --> 00:20:15,440 learning theory. 345 00:20:15,440 --> 00:20:22,790 But by the time you get to the mid-20th century and the 346 00:20:22,790 --> 00:20:27,340 heyday of behaviorism and all that good Skinnarian pigeons 347 00:20:27,340 --> 00:20:32,990 in the box kind of thing, you get a series of language 348 00:20:32,990 --> 00:20:36,910 learning that say well, look, language learning is just a 349 00:20:36,910 --> 00:20:40,580 fancy variety of any other kind of learning. 350 00:20:40,580 --> 00:20:45,520 You want your kid to speak English and so you reward him 351 00:20:45,520 --> 00:20:46,895 for speaking English. 352 00:20:46,895 --> 00:20:50,310 You shape the behavior. 353 00:20:50,310 --> 00:20:52,810 The same time you're trying to get the kid to do some sort of 354 00:20:52,810 --> 00:20:56,580 complicated thing like being toilet trained, you're also 355 00:20:56,580 --> 00:21:02,550 trying to get him to do things like saying I go to the store. 356 00:21:02,550 --> 00:21:04,550 Yesterday I went to the store. 357 00:21:07,310 --> 00:21:12,800 It's just the same thing more or less. 358 00:21:12,800 --> 00:21:15,450 One of the great efforts to make that into a single 359 00:21:15,450 --> 00:21:21,070 coherent story is B.F. Skinner's book called Verbal 360 00:21:21,070 --> 00:21:23,380 Behavior -- 361 00:21:23,380 --> 00:21:25,140 good name from a good behaviorist. 362 00:21:25,140 --> 00:21:28,570 It's just another kind of behavior. 363 00:21:28,570 --> 00:21:34,200 The start of the East coast story and the start of modern 364 00:21:34,200 --> 00:21:38,460 cognitive science in some ways is Noam Chomsky, then a young 365 00:21:38,460 --> 00:21:43,950 professor at MIT, writing a scathing review of that book, 366 00:21:43,950 --> 00:21:45,360 which still bears reading. 367 00:21:45,360 --> 00:21:48,810 It's still fun to read that, or you can go read the summary 368 00:21:48,810 --> 00:21:52,450 of it or the recap of it in Pinker's book. 369 00:21:52,450 --> 00:21:55,550 But you have to imagine this as sort of a cultural event. 370 00:21:55,550 --> 00:21:58,580 Skinner is the leading American psychologist of the 371 00:21:58,580 --> 00:22:05,740 time, and this young Noam Chomsky guy is going after him 372 00:22:05,740 --> 00:22:09,680 with a bat and both barrels blazing and any other weaponry 373 00:22:09,680 --> 00:22:10,300 you can think of. 374 00:22:10,300 --> 00:22:13,120 It is not a temperate review. 375 00:22:13,120 --> 00:22:17,170 It is insulting in many ways. 376 00:22:17,170 --> 00:22:20,870 It's an insulting piece written by a young guy about a 377 00:22:20,870 --> 00:22:25,590 very senior, established, important guy, and it starts a 378 00:22:25,590 --> 00:22:26,140 revolution. 379 00:22:26,140 --> 00:22:28,340 Anyway, I'm not going to go through the whole argument. 380 00:22:28,340 --> 00:22:37,740 But the sort of thing that Chomsky points out is that -- 381 00:22:37,740 --> 00:22:39,780 you know, Skinner's talking about schedules of 382 00:22:39,780 --> 00:22:41,180 reinforcement. 383 00:22:41,180 --> 00:22:42,650 The patterns of reinforcement for 384 00:22:42,650 --> 00:22:45,310 language just aren't there. 385 00:22:45,310 --> 00:22:47,310 This has been extensively developed 386 00:22:47,310 --> 00:22:48,090 since Chomsky's time. 387 00:22:48,090 --> 00:22:53,360 But you can imagine this as just an example. 388 00:22:53,360 --> 00:22:58,480 Imagine yourself back home as the little kid and you go up 389 00:22:58,480 --> 00:23:04,850 to your mother and you say, "Mommy, I loved you so much." 390 00:23:04,850 --> 00:23:09,130 What's the chance that she's going to say, "Bad child, bad 391 00:23:09,130 --> 00:23:15,470 child, it's I love you." By the same token, what's the 392 00:23:15,470 --> 00:23:21,790 chance that if you go up and say, as a precocious three 393 00:23:21,790 --> 00:23:27,590 year old, "Mother dear, I'm sorry to have to say this but 394 00:23:27,590 --> 00:23:31,250 it turns out that I actually loathe you and despise the 395 00:23:31,250 --> 00:23:35,280 ground on which you walk." Your mother is not going to 396 00:23:35,280 --> 00:23:38,880 say "That's a very sophisticated utterance for a 397 00:23:38,880 --> 00:23:39,670 three year old. 398 00:23:39,670 --> 00:23:43,640 Good job." You're not going to get -- 399 00:23:43,640 --> 00:23:45,070 look, that's a cartoon version. 400 00:23:45,070 --> 00:23:48,770 But when you go and look at substantial bodies of child 401 00:23:48,770 --> 00:23:51,530 language and their interaction with the grown-ups in their 402 00:23:51,530 --> 00:23:55,910 life, you don't find good evidence for a systematic 403 00:23:55,910 --> 00:24:00,380 correction of language or a systematic reinforcement for 404 00:24:00,380 --> 00:24:01,860 grammatical structure. 405 00:24:01,860 --> 00:24:06,560 You don't find any great sensitivity on the part of the 406 00:24:06,560 --> 00:24:08,750 kid to that direction when it's given. 407 00:24:08,750 --> 00:24:14,180 The sorts of things that every parent -- 408 00:24:14,180 --> 00:24:18,910 they show up in the child language literature -- but any 409 00:24:18,910 --> 00:24:24,370 parent can tell you that the kid comes up you and says, 410 00:24:24,370 --> 00:24:29,570 "Mommy, I loved you," and you say, "Oh, that's sweet, honey 411 00:24:29,570 --> 00:24:33,820 but it's really Mommy I love you." The kids says, "Yeah, 412 00:24:33,820 --> 00:24:37,520 that's right, Mommy, I loved you." They don't show an awful 413 00:24:37,520 --> 00:24:43,860 lot of sensitivity to this sort of negative support for 414 00:24:43,860 --> 00:24:45,140 what they're saying. 415 00:24:45,140 --> 00:24:47,770 How they actually figure out eventually that it's wrong is 416 00:24:47,770 --> 00:24:49,500 an interesting and important problem. 417 00:24:49,500 --> 00:24:54,300 But what is pretty clearly the case is that a story that will 418 00:24:54,300 --> 00:24:58,640 explain how it is that I persuaded my pigeon in my 419 00:24:58,640 --> 00:25:00,430 intro psych class to do ballet -- 420 00:25:00,430 --> 00:25:02,020 I think I told that story here. 421 00:25:02,020 --> 00:25:04,420 But anyway, I shaped my pigeon to do ballet. 422 00:25:04,420 --> 00:25:06,910 That was great. 423 00:25:06,910 --> 00:25:13,630 That such a story does not cash out well as a story for 424 00:25:13,630 --> 00:25:15,470 how you're trying to explain language. 425 00:25:15,470 --> 00:25:21,200 It does not seem to be possible to explain language 426 00:25:21,200 --> 00:25:25,350 in those terms. 427 00:25:25,350 --> 00:25:28,490 As noted, if you want to know more about this particular 428 00:25:28,490 --> 00:25:36,010 argument, I really would recommend Pinker's book. 429 00:25:36,010 --> 00:25:38,280 It's good stuff. 430 00:25:38,280 --> 00:25:39,300 So, what do we got? 431 00:25:39,300 --> 00:25:40,990 We got the notion that specific 432 00:25:40,990 --> 00:25:43,210 languages are learned. 433 00:25:43,210 --> 00:25:46,630 You didn't come into the world pre-wired for English, Arabic, 434 00:25:46,630 --> 00:25:48,690 Chinese or whatever. 435 00:25:48,690 --> 00:25:52,380 Language in the abstract, it's not exactly clear what it is 436 00:25:52,380 --> 00:25:57,120 that's innate, but something is innate that allows you to 437 00:25:57,120 --> 00:26:00,800 learn language. 438 00:26:00,800 --> 00:26:05,080 That something could be as specific as having, in a 439 00:26:05,080 --> 00:26:08,740 sense, the grammar written out in some sort of abstract form 440 00:26:08,740 --> 00:26:11,260 in your head and you're filling in the blanks. 441 00:26:11,260 --> 00:26:15,250 It could be something as non-specific as having a 442 00:26:15,250 --> 00:26:18,690 device that's designed to look for regularities 443 00:26:18,690 --> 00:26:21,860 in linguistic input. 444 00:26:21,860 --> 00:26:32,850 But something is in there that pre-disposes you and any 445 00:26:32,850 --> 00:26:40,590 linguistically normal human to learn language. 446 00:26:40,590 --> 00:26:44,320 In fact, it's not just something that's pre-disposing 447 00:26:44,320 --> 00:26:47,710 you to learn a spoken language. 448 00:26:47,710 --> 00:26:53,680 One of the most interesting bits of work in recent years 449 00:26:53,680 --> 00:26:57,200 in this area is the development of what really 450 00:26:57,200 --> 00:27:03,500 boils down to a brand new language growing up, as much 451 00:27:03,500 --> 00:27:07,630 as anything, out of whatever this innate endowment is. 452 00:27:07,630 --> 00:27:11,200 This is the story of Nicaraguan sign language 453 00:27:11,200 --> 00:27:13,530 before the Sandinista Revolution 454 00:27:13,530 --> 00:27:16,600 in the '70s in Nicaragua. 455 00:27:16,600 --> 00:27:23,770 I gather that deaf kids were typically raised at home and 456 00:27:23,770 --> 00:27:29,910 treated as though they were not just deaf, but not bright. 457 00:27:29,910 --> 00:27:31,160 They weren't going to learn much. 458 00:27:33,690 --> 00:27:35,200 I don't know if this is universally true, but there 459 00:27:35,200 --> 00:27:36,760 was a population of such kids. 460 00:27:36,760 --> 00:27:39,700 When the Revolution happened, one of the things that 461 00:27:39,700 --> 00:27:42,800 happened was a school for the deaf was set up in Managua, in 462 00:27:42,800 --> 00:27:45,820 the capital, and a bunch of deaf kids were brought 463 00:27:45,820 --> 00:27:49,170 together for the first time. 464 00:27:49,170 --> 00:27:53,000 What happened was that -- 465 00:27:53,000 --> 00:27:57,660 deaf kids, even in the absence of other members of the deaf 466 00:27:57,660 --> 00:28:03,970 community, will produce signs, gestures called home signs 467 00:28:03,970 --> 00:28:07,430 that indicate certain things. 468 00:28:07,430 --> 00:28:09,960 It's sort of a basic vocabulary, not a language, 469 00:28:09,960 --> 00:28:12,760 but sort of a core vocabulary that you can use to 470 00:28:12,760 --> 00:28:17,510 communicate with your hearing parents, for instance. 471 00:28:17,510 --> 00:28:22,950 What happened was that all these deaf kids got together 472 00:28:22,950 --> 00:28:26,640 and I don't know what was going on in school, but I 473 00:28:26,640 --> 00:28:30,780 think what was going on at night up in the dormitory was 474 00:28:30,780 --> 00:28:33,820 they were working something out so they could communicate 475 00:28:33,820 --> 00:28:35,030 with each other. 476 00:28:35,030 --> 00:28:41,080 They put together sort of a proto language. 477 00:28:41,080 --> 00:28:45,560 Then the next generation, the next cohort of kids come, and 478 00:28:45,560 --> 00:28:49,430 now there's a crude linguistic environment. 479 00:28:49,430 --> 00:28:52,890 What has happened over the next few cohorts of kids is 480 00:28:52,890 --> 00:28:56,540 that the next round of language learning kids have 481 00:28:56,540 --> 00:29:00,740 systematically developed a language that now has 482 00:29:00,740 --> 00:29:03,860 grammatical structures in it that look like the grammatical 483 00:29:03,860 --> 00:29:06,320 structures of other languages. 484 00:29:06,320 --> 00:29:09,150 American Sign Language, which is the common language used in 485 00:29:09,150 --> 00:29:12,150 the deaf community, the signing deaf community here, 486 00:29:12,150 --> 00:29:16,010 is a language like French, English, Chinese or whatever. 487 00:29:16,010 --> 00:29:20,460 It's not English translated into hand gesture. 488 00:29:20,460 --> 00:29:23,610 Nicaraguan Sign Language is not American Sign Language, 489 00:29:23,610 --> 00:29:27,860 it's a new sign language with its own set of vocabulary and 490 00:29:27,860 --> 00:29:30,940 its own grammatical structure. 491 00:29:30,940 --> 00:29:37,800 But that grammatical structure seems to reflect an underlying 492 00:29:37,800 --> 00:29:41,150 tendency for all languages to have certain 493 00:29:41,150 --> 00:29:42,130 structures to them. 494 00:29:42,130 --> 00:29:45,000 Let me give you one example because it's brand new -- this 495 00:29:45,000 --> 00:29:47,670 year's discovery in this realm. 496 00:29:47,670 --> 00:29:52,520 Suppose you're telling the Jack and Jill story, right. 497 00:29:52,520 --> 00:29:55,600 Jack and Jill went up the hill and then stuff happens and 498 00:29:55,600 --> 00:29:58,010 they fall down the hill, right. 499 00:29:58,010 --> 00:30:01,270 In English, the fall down part and the 500 00:30:01,270 --> 00:30:04,870 hill part are separate. 501 00:30:04,870 --> 00:30:06,760 They are obligatorily separate. 502 00:30:06,760 --> 00:30:08,410 There's the verb part and there's the 503 00:30:08,410 --> 00:30:10,330 object of that verb. 504 00:30:10,330 --> 00:30:12,960 You fall down and you fell down the hill. 505 00:30:12,960 --> 00:30:15,860 There's no reason that would need to be the case in a sign 506 00:30:15,860 --> 00:30:17,110 language, right. 507 00:30:17,110 --> 00:30:20,680 You could perfectly have a gesture that was blah that 508 00:30:20,680 --> 00:30:23,760 designates the act of falling down a hill. 509 00:30:23,760 --> 00:30:29,380 And indeed, apparently, the first generations of signers 510 00:30:29,380 --> 00:30:33,000 in this Nicaraguan school, have signs like that. 511 00:30:33,000 --> 00:30:35,680 But the new kids, the kids who came in young and are 512 00:30:35,680 --> 00:30:40,730 learning, developing this new grammatical version of sign 513 00:30:40,730 --> 00:30:44,180 language, have now separated those sings out and have a 514 00:30:44,180 --> 00:30:46,500 falling sign and a hill sign. 515 00:30:46,500 --> 00:30:48,520 I don't know that the signs look like that at all. 516 00:30:48,520 --> 00:30:50,600 The important piece of it though -- 517 00:30:50,600 --> 00:30:53,440 that's another misconception about signs is that the signs 518 00:30:53,440 --> 00:30:56,460 necessarily need to be sort of pictographic. 519 00:30:56,460 --> 00:30:58,100 There's no requirement that the hill 520 00:30:58,100 --> 00:30:59,570 sign look like a hill. 521 00:30:59,570 --> 00:31:02,580 But those of us who are complete illiterates in 522 00:31:02,580 --> 00:31:04,950 anything like a sign language it might as well be because it 523 00:31:04,950 --> 00:31:05,970 gives you the idea. 524 00:31:05,970 --> 00:31:09,310 So, they have developed instead of falling down the 525 00:31:09,310 --> 00:31:16,200 hill sign, a thing falling hill grammar that corresponds 526 00:31:16,200 --> 00:31:18,650 to the same sort of constraints that show up in 527 00:31:18,650 --> 00:31:20,000 other languages. 528 00:31:22,930 --> 00:31:27,730 This whole project has been interesting evidence for the 529 00:31:27,730 --> 00:31:34,190 powerful in which the human brain is there to learn, and 530 00:31:34,190 --> 00:31:38,090 even if necessary to create language. 531 00:31:38,090 --> 00:31:43,770 Well, that raises the interesting question of well, 532 00:31:43,770 --> 00:31:46,500 what about non-human brains. 533 00:31:46,500 --> 00:31:50,020 Do animals have language? 534 00:31:50,020 --> 00:31:55,040 This has been a question for a long, long time. 535 00:31:55,040 --> 00:31:57,060 It's been a scientific question for a 536 00:31:57,060 --> 00:32:03,880 reasonably long time. 537 00:32:03,880 --> 00:32:05,740 We're going to talk about it's important to distinguish 538 00:32:05,740 --> 00:32:10,860 between two different sorts of language. 539 00:32:10,860 --> 00:32:16,680 One would be any natural language that the animal might 540 00:32:16,680 --> 00:32:21,240 have on their own -- do animals talk to each other in 541 00:32:21,240 --> 00:32:25,320 the absence of humans intervening. 542 00:32:25,320 --> 00:32:28,230 The other question is can we teach them our language, can 543 00:32:28,230 --> 00:32:36,270 we teach them how to talk like us in some fashion. 544 00:32:36,270 --> 00:32:40,920 Within this natural and trained language dichotomy, 545 00:32:40,920 --> 00:32:44,700 then it's important to break the question of language down 546 00:32:44,700 --> 00:32:49,190 into smaller pieces because otherwise you end up in an 547 00:32:49,190 --> 00:32:52,170 endless sterile argument about well, what exactly do you mean 548 00:32:52,170 --> 00:32:54,580 by language. 549 00:32:54,580 --> 00:32:58,710 If I write a question on the final exam, because it won't 550 00:32:58,710 --> 00:33:00,220 be on the mid-term because today's topic 551 00:33:00,220 --> 00:33:02,200 is not on the mid-term. 552 00:33:02,200 --> 00:33:05,930 If I write a question on the final exam that basically says 553 00:33:05,930 --> 00:33:10,370 do animals have language, it would be perfectly possible to 554 00:33:10,370 --> 00:33:15,850 write a coherent answer for either the yes or no position. 555 00:33:15,850 --> 00:33:18,120 But that would depend on what it is 556 00:33:18,120 --> 00:33:19,220 that you mean by language. 557 00:33:19,220 --> 00:33:21,280 So let's break this down a little bit. 558 00:33:21,280 --> 00:33:25,960 For starters, do animals have something that 559 00:33:25,960 --> 00:33:28,260 we could call signals. 560 00:33:28,260 --> 00:33:34,170 Do they have sensory stimuli that are communicative in 561 00:33:34,170 --> 00:33:35,990 their intent? 562 00:33:35,990 --> 00:33:40,410 Well, the answer to that is clearly yes. 563 00:33:40,410 --> 00:33:46,540 There's no problem with figuring that out at all. 564 00:33:46,540 --> 00:33:52,710 A peacock's tail is a signal to the peahen. 565 00:33:52,710 --> 00:33:55,860 Tropical fish -- 566 00:33:55,860 --> 00:33:57,730 do tropical fish have all these beautiful 567 00:33:57,730 --> 00:33:58,650 colors all over them? 568 00:33:58,650 --> 00:34:03,000 Well, the idea is that I'm a tropical fish with, oh, my 569 00:34:03,000 --> 00:34:07,900 goodness with stripes on, and I gotta look for another 570 00:34:07,900 --> 00:34:12,930 tropical fish with stripes on, because if I mate with one 571 00:34:12,930 --> 00:34:15,410 with spots on, nothing's going to happen. 572 00:34:15,410 --> 00:34:18,770 But if I mate with one with stripes on, we get more little 573 00:34:18,770 --> 00:34:21,630 stripey fish. 574 00:34:21,630 --> 00:34:25,030 The peacock tail story is supposed to be slightly more 575 00:34:25,030 --> 00:34:25,670 complicated. 576 00:34:25,670 --> 00:34:29,420 The standard evolutionary story for peacock tails is 577 00:34:29,420 --> 00:34:33,830 that peacock tails are really stupid. 578 00:34:33,830 --> 00:34:37,800 They don't help you fly, they make you more likely to get 579 00:34:37,800 --> 00:34:39,840 chomped on by a predator and stuff. 580 00:34:39,840 --> 00:34:43,080 But a huge investment. 581 00:34:43,080 --> 00:34:47,200 What this id is the male peacock saying I am so fit, my 582 00:34:47,200 --> 00:34:51,890 genes are so good that I grew this whole thing 583 00:34:51,890 --> 00:34:55,090 just for you, honey. 584 00:34:55,090 --> 00:34:58,570 You really ought to come and check this out. 585 00:34:58,570 --> 00:35:01,130 So it's a signal. 586 00:35:01,130 --> 00:35:04,030 It's not just an awful lot of animal signals seem to be 587 00:35:04,030 --> 00:35:08,250 about sex and mating issues. 588 00:35:08,250 --> 00:35:09,390 Not all of them. 589 00:35:09,390 --> 00:35:14,590 So, Vervet monkeys, for instance, have three distinct 590 00:35:14,590 --> 00:35:17,150 alarm calls. 591 00:35:17,150 --> 00:35:19,080 How do you know they're distinct alarm calls? 592 00:35:19,080 --> 00:35:23,220 Well, you go off and record them, and then you get your -- 593 00:35:23,220 --> 00:35:25,440 not your iPod because they don't have the little ear -- 594 00:35:25,440 --> 00:35:28,190 you get your boom box or something and you play them to 595 00:35:28,190 --> 00:35:29,540 a bunch of Vervet monkeys. 596 00:35:29,540 --> 00:35:34,500 If you play call number one, all the Vervet 597 00:35:34,500 --> 00:35:36,380 monkeys jump up a tree. 598 00:35:36,380 --> 00:35:39,440 If you play call number two, all the Vervet monkeys jumps 599 00:35:39,440 --> 00:35:40,530 down a hole. 600 00:35:40,530 --> 00:35:42,440 If you play number three, they all stand up 601 00:35:42,440 --> 00:35:43,890 and look all around. 602 00:35:43,890 --> 00:35:50,470 Apparently, what this is about is if you're a Vervet monkey 603 00:35:50,470 --> 00:35:55,080 and you see a leopard, you give one scream and everybody 604 00:35:55,080 --> 00:35:58,580 jumps up the tree because the leopard can't go up the tree. 605 00:35:58,580 --> 00:36:02,280 If you see a hawk or whatever the big predator bird is who 606 00:36:02,280 --> 00:36:05,360 eats Vervet monkeys, you give a different screen, you should 607 00:36:05,360 --> 00:36:08,110 jump down the hole because the bird ain't going down the 608 00:36:08,110 --> 00:36:12,180 hole, and makes you want to just keep an eye out for them. 609 00:36:12,180 --> 00:36:14,340 So if somebody thinks they see a snake, everybody else's is 610 00:36:14,340 --> 00:36:17,630 just going to look around. 611 00:36:17,630 --> 00:36:19,820 There's sort of a three word vocabulary there. 612 00:36:19,820 --> 00:36:21,910 It also gets onto the second issue on the handout. 613 00:36:24,750 --> 00:36:26,330 Yes, it should say meaning. 614 00:36:26,330 --> 00:36:27,530 Do these signals have meaning? 615 00:36:27,530 --> 00:36:27,950 Well, sure. 616 00:36:27,950 --> 00:36:31,750 There's no point to having a signal of this sort if it 617 00:36:31,750 --> 00:36:33,720 doesn't have meaning. 618 00:36:33,720 --> 00:36:35,520 Well, you could have a signal without meaning. 619 00:36:35,520 --> 00:36:39,290 Look, I got a rock here. 620 00:36:39,290 --> 00:36:47,340 This crystal could be thought of as signaling its safireness 621 00:36:47,340 --> 00:36:48,580 or something like that. 622 00:36:48,580 --> 00:36:51,560 But no other safire gives a hoot about it. 623 00:36:51,560 --> 00:36:53,800 There's no meaning there from the point of view of the rock, 624 00:36:53,800 --> 00:36:55,040 it just is. 625 00:36:55,040 --> 00:36:59,610 Whereas, an animal's signals of the sort that we've been 626 00:36:59,610 --> 00:37:02,500 talking about clearly have communicative intent. 627 00:37:02,500 --> 00:37:04,760 There is a purpose here. 628 00:37:04,760 --> 00:37:08,300 It's not even confined just to animals it turns out. 629 00:37:08,300 --> 00:37:16,240 There are trees who respond to predator attacks by releasing 630 00:37:16,240 --> 00:37:18,990 a chemical. 631 00:37:18,990 --> 00:37:21,190 It's all quite a passive kind of thing. 632 00:37:21,190 --> 00:37:26,680 The bug chews on the leaves, that ruptures the cells which 633 00:37:26,680 --> 00:37:30,210 causes the chemical to be released, which can then be 634 00:37:30,210 --> 00:37:32,600 picked up by the next tree over. 635 00:37:32,600 --> 00:37:36,930 The next tree over in reponse now mobilizes a defense 636 00:37:36,930 --> 00:37:38,530 against that predator. 637 00:37:38,530 --> 00:37:42,500 So, in a sense, the trees are talking to each other. 638 00:37:42,500 --> 00:37:47,270 There's a signal there and it's being received. 639 00:37:47,270 --> 00:37:52,360 So, no doubt that the animal kingdom has signals. 640 00:37:52,360 --> 00:37:57,100 The equivalent in human language would be phoneme. 641 00:37:57,100 --> 00:38:02,600 No doubt that these signals have meaning, at least in some 642 00:38:02,600 --> 00:38:05,650 cases, and that could be considered to be sort of the 643 00:38:05,650 --> 00:38:07,100 equivalent of morpheme. 644 00:38:09,690 --> 00:38:10,900 What about human language? 645 00:38:10,900 --> 00:38:14,990 What about training an animal to use human language? 646 00:38:14,990 --> 00:38:21,360 Could be the case that the only reason that a chimp, who 647 00:38:21,360 --> 00:38:26,030 is a close relation of ours, hasn't talked to us is because 648 00:38:26,030 --> 00:38:28,480 nobody talked to him. 649 00:38:28,480 --> 00:38:32,790 So back in the '30s, the Hayes's, a couple in Salt Lake 650 00:38:32,790 --> 00:38:34,780 City as I recall, tried this out. 651 00:38:34,780 --> 00:38:39,370 What they did was they had a kid, [? Maracatridas. ?] 652 00:38:39,370 --> 00:38:41,660 They had a kid, they got a chimp. 653 00:38:41,660 --> 00:38:44,750 They said we'll raise the two of them identically. 654 00:38:44,750 --> 00:38:45,790 I don't know, the chimp might sleep 655 00:38:45,790 --> 00:38:47,040 through the night sooner. 656 00:38:49,410 --> 00:38:52,120 They'll talk to the kid, they'll talk to the chimp. 657 00:38:52,120 --> 00:38:56,900 After how long? 658 00:38:56,900 --> 00:38:58,410 It doesn't say how long they were at it. 659 00:38:58,410 --> 00:39:01,560 You don't want to do this for too long because after awhile, 660 00:39:01,560 --> 00:39:05,980 raising your chimp just hanging around in the play 661 00:39:05,980 --> 00:39:10,820 room with your a kid, is not going to be good for the kid, 662 00:39:10,820 --> 00:39:14,880 because the chimp may not talk a lot, but the chimp grows big 663 00:39:14,880 --> 00:39:20,150 and strong and will play with your kid in ways that your kid 664 00:39:20,150 --> 00:39:22,050 will not enjoy. 665 00:39:22,050 --> 00:39:26,870 But the chimp didn't talk. 666 00:39:26,870 --> 00:39:31,520 This is important, because any neurologically normal child 667 00:39:31,520 --> 00:39:35,930 raised in a linguistic environment will talk, will 668 00:39:35,930 --> 00:39:42,400 learn to talk, or if deaf will communicate, you and given 669 00:39:42,400 --> 00:39:46,420 appropriate input will communicate in other ways. 670 00:39:46,420 --> 00:39:54,620 The chimp is reputed to have said, I think mamma, papa and 671 00:39:54,620 --> 00:39:58,850 cup, and that was it. 672 00:39:58,850 --> 00:40:03,120 Now, one of the problems here that was quickly identified is 673 00:40:03,120 --> 00:40:06,930 that the chimp's vocal apparatus is not like yours 674 00:40:06,930 --> 00:40:11,720 and mine and is poorly designed to make the range of 675 00:40:11,720 --> 00:40:17,260 human phoneme that are important in human language. 676 00:40:17,260 --> 00:40:25,750 So, it occured to subsequent researchers to try using sign 677 00:40:25,750 --> 00:40:31,080 language with chimps because chimps are very good with 678 00:40:31,080 --> 00:40:33,310 their hands -- there's no problem making the sort of 679 00:40:33,310 --> 00:40:36,660 gestures, there's no physical limitation on the soft to 680 00:40:36,660 --> 00:40:41,490 gestures that the chimps could make or gorillas. 681 00:40:41,490 --> 00:40:46,500 Limited work with orangutans who turn out to be big and 682 00:40:46,500 --> 00:40:48,920 lazy, I understand. 683 00:40:48,920 --> 00:40:54,360 But chimps, Bonobo chimps and gorillas 684 00:40:54,360 --> 00:40:55,830 all make good signers. 685 00:40:59,050 --> 00:41:03,960 So most of the effort to teach human language to primates has 686 00:41:03,960 --> 00:41:09,160 involved sign language or computer keyboard arrangements 687 00:41:09,160 --> 00:41:13,670 where you're sort of teaching the animal to type, in a 688 00:41:13,670 --> 00:41:19,630 sense, rather than to sign. 689 00:41:19,630 --> 00:41:25,350 Now it's true that chimps don't have the vocal apparatus 690 00:41:25,350 --> 00:41:27,610 to talk the way you and I talk. 691 00:41:27,610 --> 00:41:32,120 But that couldn't be the whole story, because if you have a 692 00:41:32,120 --> 00:41:35,690 child born with something congenital malformation that 693 00:41:35,690 --> 00:41:39,640 makes it impossible for them to talk properly, that child 694 00:41:39,640 --> 00:41:44,110 will talk badly but will talk and will learn a language. 695 00:41:44,110 --> 00:41:49,050 A chimp in the same situation will simply not talk. 696 00:41:49,050 --> 00:41:53,060 There is a fundamental difference in the way that the 697 00:41:53,060 --> 00:41:55,480 chimp and the child are responding 698 00:41:55,480 --> 00:41:58,550 to linguistic input. 699 00:42:02,480 --> 00:42:06,510 So, can you teach it to sign? 700 00:42:06,510 --> 00:42:06,760 Yes. 701 00:42:06,760 --> 00:42:08,050 Actually, you can teach them to sign. 702 00:42:08,050 --> 00:42:12,560 They've got a whole string of chimps who do really quite 703 00:42:12,560 --> 00:42:20,960 well with vocabulary on the order of hundreds up signs 704 00:42:20,960 --> 00:42:25,300 that seem to have discreet meaning. 705 00:42:25,300 --> 00:42:28,960 It's not limited even to chimps. 706 00:42:28,960 --> 00:42:33,490 Irene Pepperberg who was here for a while, is now a Fellow 707 00:42:33,490 --> 00:42:38,170 at Harvard this year, has made a career out of studying 708 00:42:38,170 --> 00:42:40,430 African Grey parrots. 709 00:42:40,430 --> 00:42:43,940 It's a not terribly dramatic, nice looking little parrot 710 00:42:43,940 --> 00:42:47,660 with a nice little bird brain. 711 00:42:47,660 --> 00:42:50,690 The nice thing, of course, about parrots is that they can 712 00:42:50,690 --> 00:42:55,570 form the sounds -- they're pretty good with the phoneme, 713 00:42:55,570 --> 00:42:57,910 Polly want a cracker and all that sort of stuff. 714 00:42:57,910 --> 00:42:59,750 But they're also remarkably bright. 715 00:43:02,430 --> 00:43:08,740 Alex, her great African Grey has a vocabulary on the order 716 00:43:08,740 --> 00:43:19,050 of 50 to 100 words, I think, that he can use in various and 717 00:43:19,050 --> 00:43:20,450 reasonable ways. 718 00:43:20,450 --> 00:43:27,470 So if Irene says to Alex, pick up the blue thing or the blue 719 00:43:27,470 --> 00:43:32,230 one, Alex will toddle over and grab a blue thing. 720 00:43:32,230 --> 00:43:34,930 If you say get the truck, it'll go get the truck and 721 00:43:34,930 --> 00:43:35,600 stuff like that. 722 00:43:35,600 --> 00:43:41,690 It clearly has the grasp of vocabulary and ability to use 723 00:43:41,690 --> 00:43:47,290 it in some fashion or other. 724 00:43:47,290 --> 00:43:48,860 Mostly trained. 725 00:43:48,860 --> 00:43:52,400 Almost all of the examples of animal use of a human 726 00:43:52,400 --> 00:43:55,620 language, any aspect of a human 727 00:43:55,620 --> 00:43:57,070 language, are heavily trained. 728 00:43:57,070 --> 00:44:00,840 Again, that's different from a little kid where a little kid 729 00:44:00,840 --> 00:44:05,120 will pick up a language just by being around speakers. 730 00:44:05,120 --> 00:44:09,400 There are a couple of instances now in the 731 00:44:09,400 --> 00:44:16,680 literature where children, the offspring of signing apes have 732 00:44:16,680 --> 00:44:18,410 picked up signing on their own. 733 00:44:18,410 --> 00:44:20,920 So it may, in fact, be the case that you don't need 734 00:44:20,920 --> 00:44:23,330 explicit training. 735 00:44:23,330 --> 00:44:32,070 But what does seem to be the case is that chimps and 736 00:44:32,070 --> 00:44:38,560 parrots can use what you'd probably consider to be 737 00:44:38,560 --> 00:44:39,960 morpheme -- 738 00:44:39,960 --> 00:44:42,530 signs with meaning of some sort. 739 00:44:42,530 --> 00:44:45,170 There are issues about the data quality. 740 00:44:45,170 --> 00:44:51,360 So if you point to something, if you point to this banana 741 00:44:51,360 --> 00:44:56,510 and ask one of signing apes what it is, you'll probably 742 00:44:56,510 --> 00:45:00,170 get banana but you might get something out like banana, 743 00:45:00,170 --> 00:45:02,580 banana, me tickle, banana, you tickle me, 744 00:45:02,580 --> 00:45:04,190 banana, banana or something. 745 00:45:04,190 --> 00:45:06,100 You might wonder whether or not you should 746 00:45:06,100 --> 00:45:09,050 code that as banana. 747 00:45:12,790 --> 00:45:16,600 I saw one description that said that chimps tend to run 748 00:45:16,600 --> 00:45:18,110 on at the hands -- 749 00:45:18,110 --> 00:45:21,470 you know, the expression that somebody runs on at the mouth. 750 00:45:21,470 --> 00:45:25,700 Well, chimps sign a lot, these signing chimps, perhaps 751 00:45:25,700 --> 00:45:28,920 because they're playing the same guessing game that we're 752 00:45:28,920 --> 00:45:30,150 all playing with language. 753 00:45:30,150 --> 00:45:32,790 What does this guy want from me? 754 00:45:32,790 --> 00:45:33,900 What's he trying to -- 755 00:45:33,900 --> 00:45:35,940 pointing at that thing, oh yes. 756 00:45:35,940 --> 00:45:40,370 If I make these banana signs, maybe a few other signs in 757 00:45:40,370 --> 00:45:43,070 there, good things will happen, he'll be happy with me 758 00:45:43,070 --> 00:45:45,250 and maybe we'll get good stuff. 759 00:45:45,250 --> 00:45:51,300 So, in any case, we've got evidence for signals, we've 760 00:45:51,300 --> 00:45:53,250 got evidence that they have meaning. 761 00:45:53,250 --> 00:45:55,070 How about grammar? 762 00:45:55,070 --> 00:46:00,280 Is there any evidence for grammar in animals? 763 00:46:00,280 --> 00:46:02,770 Things get thinner there. 764 00:46:02,770 --> 00:46:09,510 Recall that you can use grammatical information to get 765 00:46:09,510 --> 00:46:12,020 yourself into the meaning of an utterance, even if you 766 00:46:12,020 --> 00:46:13,050 don't really--. 767 00:46:13,050 --> 00:46:14,630 Did I put it on the handout? 768 00:46:14,630 --> 00:46:17,870 "Colorless green ideas sleep furiously," which I think is a 769 00:46:17,870 --> 00:46:22,750 Chomsky-ism originally designed to point out that 770 00:46:22,750 --> 00:46:27,720 purely syntactic information tells you something, even 771 00:46:27,720 --> 00:46:32,010 without much in the way of, you know how that 772 00:46:32,010 --> 00:46:33,010 sentence put together. 773 00:46:33,010 --> 00:46:35,900 Actually, this goes back before that. 774 00:46:35,900 --> 00:46:38,980 How many of you are great Alice In Wonderland fans from 775 00:46:38,980 --> 00:46:41,010 your youth at some point? 776 00:46:41,010 --> 00:46:43,190 Oh, very limited. 777 00:46:43,190 --> 00:46:46,130 So, most people cannot sit here and quote back to me 778 00:46:46,130 --> 00:46:49,900 "'Twas Brillig and the slithy toves, did gyre and gimble in 779 00:46:49,900 --> 00:46:53,700 the wabe, all mimsy were the borogoves and the mome raths 780 00:46:53,700 --> 00:46:57,740 outgrabe." You have no idea what that's about. 781 00:46:57,740 --> 00:46:59,370 AUDIENCE: [INAUDIBLE]. 782 00:46:59,370 --> 00:47:02,850 PROFESSOR: It's the Jabberwocky poem in Through 783 00:47:02,850 --> 00:47:04,180 The Looking Glass? 784 00:47:04,180 --> 00:47:07,150 I think it's the Looking Glass. 785 00:47:07,150 --> 00:47:10,990 Anyway, it's in one of the two Alice books. 786 00:47:10,990 --> 00:47:13,850 But the interesting thing about it is even though you 787 00:47:13,850 --> 00:47:19,110 have no idea "'Twas Billig and the slithy toves, did gyre and 788 00:47:19,110 --> 00:47:21,900 gimble in the wabe." You, what? 789 00:47:21,900 --> 00:47:25,030 Well, but you know that now that there are some things 790 00:47:25,030 --> 00:47:26,970 cold toves. 791 00:47:26,970 --> 00:47:30,360 They have the attribute of slithiness. 792 00:47:30,360 --> 00:47:33,290 They can do things that are 793 00:47:33,290 --> 00:47:36,580 apparently gyring and gimbling. 794 00:47:36,580 --> 00:47:40,470 You can say things like gyring and gimbling -- i-n-g'ing 795 00:47:40,470 --> 00:47:43,790 these things, even though I got no idea what this verb 796 00:47:43,790 --> 00:47:45,280 means, you just know it's a verb. 797 00:47:45,280 --> 00:47:49,280 So that's the sort of power of the grammar of the syntax of 798 00:47:49,280 --> 00:47:50,270 the language. 799 00:47:50,270 --> 00:47:55,230 To what extent do animals have a grip on that? 800 00:47:55,230 --> 00:47:59,810 The answer seems to be a pretty limited grip. 801 00:47:59,810 --> 00:48:08,120 There is evidence for word order sensitivity in animal 802 00:48:08,120 --> 00:48:10,180 signing, for instance. 803 00:48:10,180 --> 00:48:21,090 So you can have the chimp equivalent of John hit Mary 804 00:48:21,090 --> 00:48:25,490 being different from Mary hit John. 805 00:48:25,490 --> 00:48:29,560 Those are the same word, different 806 00:48:29,560 --> 00:48:32,340 order, different meaning. 807 00:48:32,340 --> 00:48:34,440 Chimps will show a sensitivity to that. 808 00:48:34,440 --> 00:48:40,840 In fact, when this was a hot issue and people were arguing 809 00:48:40,840 --> 00:48:45,190 about it vociferously, Epstein up at Harvard who was one of 810 00:48:45,190 --> 00:48:47,650 Skinner's last students, used to have a great time trying to 811 00:48:47,650 --> 00:48:52,680 do these fancy chimp language demonstrations with pigeons. 812 00:48:52,680 --> 00:48:55,960 He showed that sure enough, you could get a pigeon to 813 00:48:55,960 --> 00:49:00,300 understand that you got a bunch of colored keys in your 814 00:49:00,300 --> 00:49:06,440 Skinner box, that if I pick red, green, blue I get food, 815 00:49:06,440 --> 00:49:10,320 but if I do blue, green, red I get water, and the pigeon 816 00:49:10,320 --> 00:49:12,510 would respond appropriately to that. 817 00:49:12,510 --> 00:49:18,380 So order information is there in chimp signing and it's in 818 00:49:18,380 --> 00:49:20,660 there in other behaviors too. 819 00:49:20,660 --> 00:49:24,910 There's order information, structural information in 820 00:49:24,910 --> 00:49:27,640 natural animal language, too. 821 00:49:27,640 --> 00:49:33,200 It's not just ah means snake or something like that. 822 00:49:33,200 --> 00:49:36,390 So, for example, bees. 823 00:49:36,390 --> 00:49:45,280 How many of you have seen the waggle dance of bees? 824 00:49:45,280 --> 00:49:47,440 It's pretty thin stuff here. 825 00:49:47,440 --> 00:49:50,160 You get in free to the science museum here, I 826 00:49:50,160 --> 00:49:52,490 believe, as MIT students. 827 00:49:52,490 --> 00:49:57,110 Sometime, like Friday when you're done with the mid-term 828 00:49:57,110 --> 00:49:59,880 or something, you want to go over to the science museum, 829 00:49:59,880 --> 00:50:01,920 you go up to the third floor, fourth floor 830 00:50:01,920 --> 00:50:02,775 where the library is. 831 00:50:02,775 --> 00:50:04,380 They got a beehive there. 832 00:50:04,380 --> 00:50:05,750 If you look at this beehive -- 833 00:50:05,750 --> 00:50:09,330 it's a big glass plane so you can look in on it -- what you 834 00:50:09,330 --> 00:50:11,730 will see, not so much at the time of year because they're 835 00:50:11,730 --> 00:50:12,890 all going to bed at this point. 836 00:50:12,890 --> 00:50:19,620 But anyway, what you'll see is a bee who will be marching up 837 00:50:19,620 --> 00:50:28,830 the hive with a characteristic waggle of her backside, and 838 00:50:28,830 --> 00:50:35,220 then looping back down and repeating this while a whole 839 00:50:35,220 --> 00:50:37,750 bunch of other bees are gathered around -- it's easy 840 00:50:37,750 --> 00:50:42,650 to find because typically there's sort of a circle of 841 00:50:42,650 --> 00:50:45,250 bees watching this. 842 00:50:45,250 --> 00:50:49,760 This is a highly structured bit of communication about 843 00:50:49,760 --> 00:50:53,340 where to go to look for food. 844 00:50:53,340 --> 00:50:58,040 I didn't go off and review my waggle dance information, but 845 00:50:58,040 --> 00:51:03,570 I'm pretty sure that memory serves correctly to say that 846 00:51:03,570 --> 00:51:08,210 the length of the vector here gives information about how 847 00:51:08,210 --> 00:51:10,190 far you've got to go. 848 00:51:10,190 --> 00:51:13,410 That the direction of the vector is giving you 849 00:51:13,410 --> 00:51:16,910 orientation relative to the sun about which 850 00:51:16,910 --> 00:51:18,920 direction to fly. 851 00:51:18,920 --> 00:51:22,780 I think amplitude of the waggle is telling you about 852 00:51:22,780 --> 00:51:27,140 the magnitude of the food source -- how 853 00:51:27,140 --> 00:51:28,480 good is this stuff. 854 00:51:31,110 --> 00:51:32,720 Got any waggle dance experts? 855 00:51:32,720 --> 00:51:35,810 AUDIENCE: [INAUDIBLE PHRASE]. 856 00:51:35,810 --> 00:51:36,780 PROFESSOR: An altitude 857 00:51:36,780 --> 00:51:38,030 AUDIENCE: [INAUDIBLE PHRASE]. 858 00:51:41,180 --> 00:51:43,740 PROFESSOR: Signaling how high you should fly? 859 00:51:43,740 --> 00:51:46,260 AUDIENCE: [INAUDIBLE PHRASE]. 860 00:51:46,260 --> 00:51:46,880 PROFESSOR: That's cool. 861 00:51:46,880 --> 00:51:48,010 I didn't know that. 862 00:51:48,010 --> 00:51:50,650 Why would they care? 863 00:51:50,650 --> 00:51:52,280 AUDIENCE: [INAUDIBLE PHRASE]. 864 00:51:52,280 --> 00:51:55,700 PROFESSOR: Oh, of course. 865 00:51:55,700 --> 00:51:58,670 Not all the flowers are just lying on the ground. 866 00:51:58,670 --> 00:52:01,710 Particularly in Boston, it could be a roof garden. 867 00:52:01,710 --> 00:52:02,920 In any case, that's cool. 868 00:52:02,920 --> 00:52:04,360 But you don't remember what the signal is? 869 00:52:04,360 --> 00:52:04,950 AUDIENCE: [INAUDIBLE]. 870 00:52:04,950 --> 00:52:08,890 AUDIENCE: The looping back piece contains information 871 00:52:08,890 --> 00:52:10,260 too, but I don't remember what that is. 872 00:52:10,260 --> 00:52:13,750 Very clever research that people do one this. 873 00:52:13,750 --> 00:52:17,290 It's a lovely body of literature. 874 00:52:17,290 --> 00:52:24,590 But in any case, very structured and full of a sort 875 00:52:24,590 --> 00:52:29,780 of native grammar. 876 00:52:29,780 --> 00:52:34,560 But it's pretty limited stuff. 877 00:52:38,740 --> 00:52:43,720 The newest evidence for what the limitation might be that 878 00:52:43,720 --> 00:52:48,340 makes animal grammar fundamentally different from 879 00:52:48,340 --> 00:52:53,560 your grammar and my grammar comes from an experiment by -- 880 00:52:53,560 --> 00:52:54,950 that I think I put on the handout. 881 00:52:54,950 --> 00:52:58,130 Yeah, new paper by Fitch and Hauser where they basically 882 00:52:58,130 --> 00:53:04,690 were teaching grammatical structures to primates, to 883 00:53:04,690 --> 00:53:06,570 non-human the primates. 884 00:53:10,130 --> 00:53:11,050 They were pretty good. 885 00:53:11,050 --> 00:53:14,980 They were doing fine on the John hit Mary is different 886 00:53:14,980 --> 00:53:16,870 from the Mary hit John part. 887 00:53:16,870 --> 00:53:20,450 But it was nested construction that they simply 888 00:53:20,450 --> 00:53:23,460 never managed to get. 889 00:53:23,460 --> 00:53:31,350 So, Mary who was engaged to Sam hit John. 890 00:53:31,350 --> 00:53:35,030 That's perfectly fine for you, sticking one 891 00:53:35,030 --> 00:53:37,200 clause inside another. 892 00:53:37,200 --> 00:53:41,490 That was stuff that they could just never manage to get their 893 00:53:41,490 --> 00:53:44,930 primates to pick up on. 894 00:53:44,930 --> 00:53:46,760 Now this is not an argument -- 895 00:53:46,760 --> 00:53:49,720 Hauser in particular is not somebody who spends a lot of 896 00:53:49,720 --> 00:53:56,120 time arguing that apes are stupid or something like that. 897 00:53:56,120 --> 00:54:00,750 The great apes are remarkably clever beasts. 898 00:54:00,750 --> 00:54:06,900 In studying them you find cognitive abilities not unlike 899 00:54:06,900 --> 00:54:09,390 a lot of our cognitive abilities. 900 00:54:09,390 --> 00:54:11,160 I don't think it's a Hauser experiment, but one of the 901 00:54:11,160 --> 00:54:14,530 ones that comes to mind, that I could have talked about in 902 00:54:14,530 --> 00:54:17,780 the context of egocentric behavior before, is that 903 00:54:17,780 --> 00:54:25,640 important landmark in your mental development was when 904 00:54:25,640 --> 00:54:27,160 you learned how to lie. 905 00:54:29,940 --> 00:54:32,820 It's another one of these ones that didn't necessarily get a 906 00:54:32,820 --> 00:54:35,690 lot of positive reinforcement from your parents when you 907 00:54:35,690 --> 00:54:37,050 demonstrated it to them. 908 00:54:37,050 --> 00:54:39,670 But it is an important cognitive landmark because it 909 00:54:39,670 --> 00:54:45,050 says I know that what's in here is not what's in there, 910 00:54:45,050 --> 00:54:46,170 in your head. 911 00:54:46,170 --> 00:54:49,950 I could tell you something that I know is bogus and 912 00:54:49,950 --> 00:54:52,260 you'll believe it. 913 00:54:52,260 --> 00:54:54,550 Monkeys can do that too. 914 00:54:54,550 --> 00:54:55,680 How do you know this? 915 00:54:55,680 --> 00:54:58,770 Well, I can't remember who did the experiment, but he set up 916 00:54:58,770 --> 00:55:07,450 this clever experiment with a good person and a bad person. 917 00:55:07,450 --> 00:55:11,830 The monkey knows both of these people. 918 00:55:11,830 --> 00:55:17,030 Both of these people have access to 919 00:55:17,030 --> 00:55:22,130 these boxes of stuff. 920 00:55:22,130 --> 00:55:24,770 The monkey sometimes has access to the box of stuff. 921 00:55:24,770 --> 00:55:27,950 When the monkey gets access, he can look under the box and 922 00:55:27,950 --> 00:55:29,080 he finds the banana. 923 00:55:29,080 --> 00:55:29,670 Oh, cool. 924 00:55:29,670 --> 00:55:30,180 Great. 925 00:55:30,180 --> 00:55:32,040 Or the grapes or there's nothing under 926 00:55:32,040 --> 00:55:33,710 there, that's boring. 927 00:55:33,710 --> 00:55:36,675 Now the monkey's on the other side of the cage 928 00:55:36,675 --> 00:55:38,930 and here's the stuff. 929 00:55:38,930 --> 00:55:42,200 If the good guy comes in, he looks under one of these 930 00:55:42,200 --> 00:55:47,110 things, finds the banana and he shares it with the monkey. 931 00:55:47,110 --> 00:55:50,970 If the bad guy comes in, looks, finds the banana, he 932 00:55:50,970 --> 00:55:54,690 eats it and doesn't give the monkey any. 933 00:55:54,690 --> 00:55:59,920 So now, do this a few times -- monkey's got this bit down. 934 00:55:59,920 --> 00:56:05,440 Now a third party comes in and baits one of 935 00:56:05,440 --> 00:56:07,070 these two wells -- 936 00:56:07,070 --> 00:56:09,920 puts the banana under one of them. 937 00:56:09,920 --> 00:56:15,450 Now if the good guy comes in what does the monkey do? 938 00:56:15,450 --> 00:56:17,910 The monkey basically says look there. 939 00:56:17,910 --> 00:56:20,970 You only get the look there, that's where you want to look. 940 00:56:20,970 --> 00:56:23,270 The bad guy comes in, what's the monkey do? 941 00:56:23,270 --> 00:56:27,800 Let's look there. 942 00:56:27,800 --> 00:56:30,010 Monkeys will also do this to each other, 943 00:56:30,010 --> 00:56:32,140 apparently, in the wild. 944 00:56:32,140 --> 00:56:39,120 There's a version where you stash stuff around a field and 945 00:56:39,120 --> 00:56:41,070 -- now how does it work that the 946 00:56:41,070 --> 00:56:44,770 monkey lies to his friends? 947 00:56:44,770 --> 00:56:48,900 You arrange for monkey one to discover the good stash. 948 00:56:48,900 --> 00:56:52,080 Then this monkey in some fashion goes off and tells his 949 00:56:52,080 --> 00:56:54,030 buddies, go look over there -- the good stuff's over here. 950 00:56:54,030 --> 00:56:55,870 The monkey's, you guys go over there. 951 00:56:55,870 --> 00:56:57,470 And he goes and eats all the bananas. 952 00:56:57,470 --> 00:57:04,160 So, they're clever animals, but there do seem to be 953 00:57:04,160 --> 00:57:08,450 fundamental differences that you don't get around. 954 00:57:08,450 --> 00:57:11,120 At least nobody's figured out how to get around them yet. 955 00:57:11,120 --> 00:57:16,870 Nobody has figured a way how to get monkeys to anything 956 00:57:16,870 --> 00:57:20,400 like the richness of the grammar that 957 00:57:20,400 --> 00:57:21,800 human language has. 958 00:57:21,800 --> 00:57:24,980 The result of this is that even if you want to concede -- 959 00:57:24,980 --> 00:57:27,210 this is why you can answer both ways to the do animals 960 00:57:27,210 --> 00:57:28,080 have language. 961 00:57:28,080 --> 00:57:32,020 Suppose you decide that all you want is some evidence for 962 00:57:32,020 --> 00:57:35,730 phonemes, morphemes and an ability to be 963 00:57:35,730 --> 00:57:37,120 sensitive to structure. 964 00:57:37,120 --> 00:57:40,220 Well, yeah, there's some evidence that chimps can do 965 00:57:40,220 --> 00:57:41,470 that, for instance. 966 00:57:41,470 --> 00:57:46,390 But the hurdle of developing a productive language, a 967 00:57:46,390 --> 00:57:52,560 language where you can say new stuff, that does not seem to 968 00:57:52,560 --> 00:58:01,120 be a hurdle that any non-human species gets over in either 969 00:58:01,120 --> 00:58:06,790 natural language or in trained language. 970 00:58:06,790 --> 00:58:10,930 So, bees -- great sophisticated stuff, but no 971 00:58:10,930 --> 00:58:13,330 evidence that the bee ever learns to talk about anything 972 00:58:13,330 --> 00:58:15,850 new with this. 973 00:58:15,850 --> 00:58:18,210 The bee doesn't get to -- they don't get to sit around and 974 00:58:18,210 --> 00:58:22,620 apparently muse about, remember last year? 975 00:58:22,620 --> 00:58:28,110 Remember that patch of flowers that was over there? 976 00:58:28,110 --> 00:58:30,970 At least nobody to my knowledge has ever found the 977 00:58:30,970 --> 00:58:33,920 past tense in waggle or anything. 978 00:58:33,920 --> 00:58:39,860 It's their up to a limit, but not further. 979 00:58:39,860 --> 00:58:45,240 Now, what I will do is take a momentary break here, say a 980 00:58:45,240 --> 00:58:49,490 word about why does this matter, and then what I'll do 981 00:58:49,490 --> 00:58:53,850 is entertain mid-term questions for a few minutes 982 00:58:53,850 --> 00:58:56,430 before the end of the hour. 983 00:58:56,430 --> 00:58:59,680 So, take a quick stretch and then let's come back and 984 00:58:59,680 --> 00:59:00,930 finish things up here. 985 00:59:10,350 --> 00:59:14,380 Why does it matter? 986 00:59:14,380 --> 00:59:24,660 This has been a very contentious topic, moreso than 987 00:59:24,660 --> 00:59:27,820 many -- people argue about visual attention, but people 988 00:59:27,820 --> 00:59:31,240 don't get really agitated about it like 989 00:59:31,240 --> 00:59:32,910 it's a moral issue. 990 00:59:32,910 --> 00:59:37,470 People get agitated about animal language like it is a 991 00:59:37,470 --> 00:59:42,440 moral issue, because in some fashion it is -- 992 00:59:42,440 --> 00:59:46,480 it is not itself a moral issue, but it impinges on 993 00:59:46,480 --> 00:59:48,180 issues that have more to do with 994 00:59:48,180 --> 00:59:50,850 morality than with science. 995 00:59:50,850 --> 00:59:52,410 Well, you could couch it in theological 996 00:59:52,410 --> 00:59:53,550 terms, if you like. 997 00:59:53,550 --> 00:59:59,590 In Psalms, the psalmist asks rhetorically at some point of 998 00:59:59,590 --> 01:00:04,620 God, What is man that you are mindful of him? 999 01:00:04,620 --> 01:00:11,050 What is it that makes us so special that we're different 1000 01:00:11,050 --> 01:00:15,170 from all these animal things out there. 1001 01:00:17,750 --> 01:00:23,370 That's the question that becomes difficult if there is 1002 01:00:23,370 --> 01:00:29,120 not at nice clean, qualitative difference between what an 1003 01:00:29,120 --> 01:00:32,740 animal is and what us is. 1004 01:00:32,740 --> 01:00:35,950 This was one of the reasons why the advent of the theory 1005 01:00:35,950 --> 01:00:42,900 of evolution was so problematic because it 1006 01:00:42,900 --> 01:00:48,580 suggested that species were not completely distinct and 1007 01:00:48,580 --> 01:00:50,610 that we were somehow a completely separate sort of 1008 01:00:50,610 --> 01:00:54,780 creation, but that there was a continuity between all forms 1009 01:00:54,780 --> 01:00:59,280 of animal life up to and including us. 1010 01:00:59,280 --> 01:01:06,660 If that's the case, who is it that gets the sort of rights 1011 01:01:06,660 --> 01:01:10,740 and privileges that we would arrogate to ourselves, and who 1012 01:01:10,740 --> 01:01:15,690 is it that we get to treat like animals? 1013 01:01:15,690 --> 01:01:17,830 The search for a nice clear -- 1014 01:01:17,830 --> 01:01:19,510 I'm not going to give you the answer to this because there 1015 01:01:19,510 --> 01:01:23,720 is, at least in my view, there is no nice sharp, dividing 1016 01:01:23,720 --> 01:01:26,350 line that gives you a nice clean answer to this. 1017 01:01:26,350 --> 01:01:30,300 But to search for a nice clean answer could include, well, 1018 01:01:30,300 --> 01:01:32,000 who can talk to us? 1019 01:01:32,000 --> 01:01:37,360 Maybe the ability to speak or communicate in a linguistic 1020 01:01:37,360 --> 01:01:39,220 way is the critical division. 1021 01:01:39,220 --> 01:01:41,030 Because I mean look, think about it in terms of 1022 01:01:41,030 --> 01:01:43,260 what you would eat. 1023 01:01:43,260 --> 01:01:49,810 If it were the case that you'll walk out into the field 1024 01:01:49,810 --> 01:01:54,560 and the cow can talk to you and carry on a conversation, 1025 01:01:54,560 --> 01:01:57,620 even sort of a simple conversation, a child-like 1026 01:01:57,620 --> 01:01:58,265 conversation. 1027 01:01:58,265 --> 01:02:02,580 Good grass, like grass, nice grass. 1028 01:02:02,580 --> 01:02:05,210 What's the chance that you're going to eat him? 1029 01:02:05,210 --> 01:02:09,510 It doesn't feel quite right anymore. 1030 01:02:09,510 --> 01:02:10,950 It's not just language. 1031 01:02:10,950 --> 01:02:13,530 We don't like to eat things that we get to know. 1032 01:02:13,530 --> 01:02:17,820 Language is a good way to get to know something. 1033 01:02:17,820 --> 01:02:21,260 Third grade at the moment reading Charlotte's Web. 1034 01:02:21,260 --> 01:02:23,210 How many of you read Charlotte's Web at some point? 1035 01:02:23,210 --> 01:02:26,100 Oh, good, it's still a classic out there. 1036 01:02:26,100 --> 01:02:28,040 It's an issue about language and who you're 1037 01:02:28,040 --> 01:02:29,830 going to get to eat. 1038 01:02:29,830 --> 01:02:36,650 There's a great one in the little Jewish kid repertoire 1039 01:02:36,650 --> 01:02:42,130 called The Carp in the Bathtub where a big -- a Carp is 1040 01:02:42,130 --> 01:02:45,140 basically an overgrown Goldfish -- thing comes home 1041 01:02:45,140 --> 01:02:48,730 and is living in the bathtub and the kids love it but this 1042 01:02:48,730 --> 01:02:52,390 thing is intended to be food, and sad things 1043 01:02:52,390 --> 01:02:53,820 happen to the Carp. 1044 01:02:56,430 --> 01:02:58,580 Things that are close to you are not things that you want 1045 01:02:58,580 --> 01:03:02,910 to kill, not things that you necessarily want to hurt. 1046 01:03:02,910 --> 01:03:10,430 So, how you draw that line is important. 1047 01:03:10,430 --> 01:03:14,320 If you could argue that it is absolutely the case that 1048 01:03:14,320 --> 01:03:18,250 there's human language ability and animals just 1049 01:03:18,250 --> 01:03:19,570 don't got it at all--. 1050 01:03:19,570 --> 01:03:22,170 Another realm where this was tried was humans are tool 1051 01:03:22,170 --> 01:03:24,110 users and animals just aren't. 1052 01:03:24,110 --> 01:03:25,600 There's a sharp division there. 1053 01:03:25,600 --> 01:03:26,690 We can see it. 1054 01:03:26,690 --> 01:03:29,990 It makes these sorts of discussions somewhat easier. 1055 01:03:29,990 --> 01:03:32,610 It doesn't make them a lot easier necessarily. 1056 01:03:32,610 --> 01:03:37,880 Because look, if you have a human child who has a 1057 01:03:37,880 --> 01:03:41,290 congenital neurological problem that means that that 1058 01:03:41,290 --> 01:03:44,010 child will never speak -- 1059 01:03:44,010 --> 01:03:45,730 the kid's mentally retarded in some fashion, is 1060 01:03:45,730 --> 01:03:46,600 never going to speak. 1061 01:03:46,600 --> 01:03:49,800 That doesn't mean you can eat him or anything of the sort. 1062 01:03:49,800 --> 01:03:52,710 That's obvious sick nonsense. 1063 01:03:52,710 --> 01:03:56,330 So it's not a simple division. 1064 01:03:56,330 --> 01:04:03,600 But the efforts to figure out what makes us distinctively 1065 01:04:03,600 --> 01:04:12,390 human, and as a result, who do human rights extend to is a 1066 01:04:12,390 --> 01:04:16,630 serious problem, one that I'm not going to attempt to solve 1067 01:04:16,630 --> 01:04:22,070 here, but one that is certainly worth your ongoing 1068 01:04:22,070 --> 01:04:23,960 consideration. 1069 01:04:23,960 --> 01:04:28,010 What I will attempt to solve here for the remaining 10, 15 1070 01:04:28,010 --> 01:04:31,780 minutes or so, is I will dutifully attempt to answer 1071 01:04:31,780 --> 01:04:36,090 any questions about the mid-term that you might have 1072 01:04:36,090 --> 01:04:37,760 at the present moment. 1073 01:04:45,060 --> 01:04:50,260 A couple of people asked me, is it a hard mid-term? 1074 01:04:50,260 --> 01:04:55,950 The answer is that the first two TAs who read it, said boy, 1075 01:04:55,950 --> 01:04:58,190 this is easy. 1076 01:04:58,190 --> 01:05:03,810 The second two TAs who read it, said boy, this is hard. 1077 01:05:03,810 --> 01:05:07,715 Sadly, the distinction between the first two TAs and the 1078 01:05:07,715 --> 01:05:10,820 second two TAs in this case was the first two TAs had 1079 01:05:10,820 --> 01:05:14,670 TA'ed the course last year and had been through this once 1080 01:05:14,670 --> 01:05:16,890 before and the other ones hadn't. 1081 01:05:16,890 --> 01:05:19,570 So, you may use that tidbit of information in 1082 01:05:19,570 --> 01:05:21,450 any way you see fit. 1083 01:05:21,450 --> 01:05:23,110 What other tidbits of information can 1084 01:05:23,110 --> 01:05:24,360 I provide for you? 1085 01:05:27,400 --> 01:05:29,210 AUDIENCE: [INAUDIBLE PHRASE]. 1086 01:05:29,210 --> 01:05:33,560 PROFESSOR: The closed class and open class words, it's on 1087 01:05:33,560 --> 01:05:34,520 today's handout. 1088 01:05:34,520 --> 01:05:37,700 I think I mentioned this last time, but let me reiterate 1089 01:05:37,700 --> 01:05:42,070 that closed class, so it was on last time so it's fair game 1090 01:05:42,070 --> 01:05:44,270 for the mid-term. 1091 01:05:44,270 --> 01:05:51,500 Closed class words are words that serve grammaticals and 1092 01:05:51,500 --> 01:05:54,760 tactical roles and do not -- 1093 01:05:54,760 --> 01:05:58,820 the, and, uh -- 1094 01:05:58,820 --> 01:06:01,410 words that don't have a definition in any 1095 01:06:01,410 --> 01:06:05,390 straightforward sense, except to explain that the means noun 1096 01:06:05,390 --> 01:06:06,300 is a-comin'. 1097 01:06:06,300 --> 01:06:09,150 And means I'm joining a couple of things together. 1098 01:06:12,300 --> 01:06:14,510 Open class words are the other ones. 1099 01:06:14,510 --> 01:06:16,970 Closed class, it's a much smaller set. 1100 01:06:16,970 --> 01:06:22,290 The closed class words are a very small set and the open 1101 01:06:22,290 --> 01:06:25,950 class set is the set of all other words that fill up the 1102 01:06:25,950 --> 01:06:27,200 dictionary and everything else. 1103 01:06:29,560 --> 01:06:32,180 So I think that's what you need to know about that. 1104 01:06:32,180 --> 01:06:34,050 Oh, the other thing to know about it is the closed class 1105 01:06:34,050 --> 01:06:36,930 words are what shows up when you get beyond the two word 1106 01:06:36,930 --> 01:06:40,030 stage in language development. 1107 01:06:40,030 --> 01:06:45,550 Little kids, two word stage kids don't say "the ball." 1108 01:06:45,550 --> 01:06:48,330 They say "Go ball" or something like that. 1109 01:06:48,330 --> 01:06:55,660 But when they start to say "I see the cat," longer 1110 01:06:55,660 --> 01:06:57,920 utterances, that's when you start getting these closed 1111 01:06:57,920 --> 01:06:59,490 class things coming in. 1112 01:06:59,490 --> 01:07:02,640 OK, what else you need to know? 1113 01:07:02,640 --> 01:07:04,730 Everybody knows everything, this is great. 1114 01:07:04,730 --> 01:07:04,980 What? 1115 01:07:04,980 --> 01:07:06,230 AUDIENCE: [INAUDIBLE PHRASE]. 1116 01:07:12,100 --> 01:07:15,370 PROFESSOR: If you look at the exam from years past, the mix 1117 01:07:15,370 --> 01:07:16,440 roughly comparable. 1118 01:07:16,440 --> 01:07:21,140 I didn't actually check. 1119 01:07:21,140 --> 01:07:24,740 There's sort of an interesting push-pull here. 1120 01:07:24,740 --> 01:07:28,080 The longer answers are worth 10 points, so if I write long 1121 01:07:28,080 --> 01:07:34,480 answer questions, I can get done writing the exam faster. 1122 01:07:34,480 --> 01:07:35,980 I don't have to write as many questions. 1123 01:07:35,980 --> 01:07:38,870 The short answer questions, the multiple choice questions 1124 01:07:38,870 --> 01:07:40,640 are a lot easier to grade. 1125 01:07:40,640 --> 01:07:44,430 So if I write a lot of those it's an easier test to grade, 1126 01:07:44,430 --> 01:07:51,020 and so the mix of those two pressures leads me to sort of 1127 01:07:51,020 --> 01:07:53,460 the usual mix, I think. 1128 01:07:53,460 --> 01:07:55,050 It didn't look weird, did it? 1129 01:07:55,050 --> 01:07:56,490 No. 1130 01:07:56,490 --> 01:07:58,620 Kind of the usual, some of each. 1131 01:08:02,600 --> 01:08:03,850 AUDIENCE: [INAUDIBLE PHRASE]? 1132 01:08:06,440 --> 01:08:08,820 PROFESSOR: The ones of you who come here will. 1133 01:08:08,820 --> 01:08:10,650 The ones who sleep late -- 1134 01:08:10,650 --> 01:08:11,730 at least it's taught in the afternoon here. 1135 01:08:11,730 --> 01:08:14,940 Most of the people make it awake. 1136 01:08:14,940 --> 01:08:16,750 Is the final in the morning or in the afternoon? 1137 01:08:16,750 --> 01:08:17,850 AUDIENCE: [INAUDIBLE]. 1138 01:08:17,850 --> 01:08:19,010 PROFESSOR: Morning? 1139 01:08:19,010 --> 01:08:19,710 Afternoon. 1140 01:08:19,710 --> 01:08:20,980 Oh good. 1141 01:08:20,980 --> 01:08:23,200 Whenever it's in the morning there's all these sad people 1142 01:08:23,200 --> 01:08:26,340 who traipse in an hour and a half late with some great 1143 01:08:26,340 --> 01:08:29,800 story about I woke up when my alarm clock woke me up but 1144 01:08:29,800 --> 01:08:32,840 then I went back to sleep and I had a dream that I was awake 1145 01:08:32,840 --> 01:08:34,110 taking the exam. 1146 01:08:34,110 --> 01:08:39,300 Can't I get credit for the answers I was writing? 1147 01:08:39,300 --> 01:08:40,490 Yeah, it's here. 1148 01:08:40,490 --> 01:08:40,830 It's here. 1149 01:08:40,830 --> 01:08:42,360 Everybody's taking it here. 1150 01:08:42,360 --> 01:08:45,960 Look, do me a favor, do your best to put a seat between 1151 01:08:45,960 --> 01:08:49,750 each of you, and if you're not putting a seat between each of 1152 01:08:49,750 --> 01:08:50,890 you because we've run out -- 1153 01:08:50,890 --> 01:08:54,750 because there are 300 of you and not 600 seats. 1154 01:08:54,750 --> 01:09:00,260 It's considered good form not to take the exam like this, 1155 01:09:00,260 --> 01:09:02,380 particularly since the guy sitting next you is probably 1156 01:09:02,380 --> 01:09:05,700 just as clueless as you are. 1157 01:09:05,700 --> 01:09:11,020 It's really, really embarrassing when your wrong 1158 01:09:11,020 --> 01:09:13,910 answer is exactly the same as the wrong answer 1159 01:09:13,910 --> 01:09:15,900 sitting next to you. 1160 01:09:15,900 --> 01:09:18,530 I mean it's OK on a multiple choice thing, but this 1161 01:09:18,530 --> 01:09:21,670 happened -- actually, this happened to me once upon a 1162 01:09:21,670 --> 01:09:23,340 time in grade school. 1163 01:09:23,340 --> 01:09:27,280 I was accused of plagiarizing a report and they 1164 01:09:27,280 --> 01:09:30,380 showed it to me. 1165 01:09:30,380 --> 01:09:31,650 Here's this kid's report. 1166 01:09:31,650 --> 01:09:32,770 Here's your report. 1167 01:09:32,770 --> 01:09:34,450 These are identical. 1168 01:09:34,450 --> 01:09:36,130 Which was true. 1169 01:09:36,130 --> 01:09:40,900 And arrogant, smart kid that I was, I said, do you think I'd 1170 01:09:40,900 --> 01:09:44,270 plagiarize from him? 1171 01:09:44,270 --> 01:09:46,840 It didn't take much of a brain to figure out who'd plagiarize 1172 01:09:46,840 --> 01:09:50,300 from whom because this was a kid who was still reading the 1173 01:09:50,300 --> 01:09:52,960 books with the print that you could read across the room. 1174 01:09:55,870 --> 01:09:58,840 I wasn't going to plagiarize his -- 1175 01:09:58,840 --> 01:10:01,610 I was sort of flattered that he took my paper I guess. 1176 01:10:01,610 --> 01:10:04,550 Does anybody actually have a question I should answer here? 1177 01:10:04,550 --> 01:10:05,130 Yeah, all right. 1178 01:10:05,130 --> 01:10:07,660 AUDIENCE: [INAUDIBLE PHRASE]? 1179 01:10:07,660 --> 01:10:08,580 PROFESSOR: Can I go over who? 1180 01:10:08,580 --> 01:10:10,320 AUDIENCE: [INAUDIBLE PHRASE]. 1181 01:10:10,320 --> 01:10:11,770 PROFESSOR: The logic of perception. 1182 01:10:11,770 --> 01:10:17,520 Well, the logic of perception is Irvin Rock's jargon for 1183 01:10:17,520 --> 01:10:20,050 what I was really talking about in lecture, for 1184 01:10:20,050 --> 01:10:23,940 instance, as inference. 1185 01:10:23,940 --> 01:10:26,540 I think maybe in the book also is Helmoltz's idea of 1186 01:10:26,540 --> 01:10:27,710 unconscious inference. 1187 01:10:27,710 --> 01:10:32,710 It's this idea that you're continuously making the best 1188 01:10:32,710 --> 01:10:37,830 guess you can from an impoverished stimulus. 1189 01:10:37,830 --> 01:10:45,270 That you've got a set of rules built into that help you to 1190 01:10:45,270 --> 01:10:50,240 try to figure out what it is that you're looking at. 1191 01:10:50,240 --> 01:10:54,840 There's more to Rock's argument than that, but not 1192 01:10:54,840 --> 01:10:57,080 much more in the book, as I recall. 1193 01:10:57,080 --> 01:10:58,820 Something about like that. 1194 01:10:58,820 --> 01:11:01,400 There was another hand over there a moment ago. 1195 01:11:01,400 --> 01:11:02,590 Well, now there's one here. 1196 01:11:02,590 --> 01:11:03,840 AUDIENCE: [INAUDIBLE PHRASE]? 1197 01:11:08,520 --> 01:11:10,810 PROFESSOR: Lectures or books. 1198 01:11:10,810 --> 01:11:13,130 This is the triage question. 1199 01:11:13,130 --> 01:11:15,820 I haven't read the book, I wasn't at the lectures, now 1200 01:11:15,820 --> 01:11:19,150 what do I do? 1201 01:11:19,150 --> 01:11:24,630 The answer is that the intersection of book and 1202 01:11:24,630 --> 01:11:28,510 lecture is the best place to devote your energy. 1203 01:11:28,510 --> 01:11:31,460 So things that were covered in both are probably important 1204 01:11:31,460 --> 01:11:33,370 enough that -- well, they're important enough that Gleitman 1205 01:11:33,370 --> 01:11:34,860 thought it was important and I thought it was important. 1206 01:11:34,860 --> 01:11:37,440 That sounds important. 1207 01:11:37,440 --> 01:11:45,060 Beyond that, I used to say that because I am as 1208 01:11:45,060 --> 01:11:48,430 egocentric as the next person, that things that are in my 1209 01:11:48,430 --> 01:11:51,180 lectures but not the book are more likely to show up than 1210 01:11:51,180 --> 01:11:53,320 things that are in the book but not in the lecture. 1211 01:11:53,320 --> 01:11:58,800 But I'm not really sure that's true anymore. 1212 01:11:58,800 --> 01:12:02,750 So I would say that after that, if I was trying to 1213 01:12:02,750 --> 01:12:07,500 figure out how to ration my precious time, I would look at 1214 01:12:07,500 --> 01:12:11,490 the set of kind of bullet questions that I put on most 1215 01:12:11,490 --> 01:12:13,890 of the handouts, until I ran out of time, 1216 01:12:13,890 --> 01:12:16,310 for the chapter readings. 1217 01:12:16,310 --> 01:12:19,730 Because what you've got there is what I think is important 1218 01:12:19,730 --> 01:12:24,310 in the chapter, and that's a pretty good bet on things that 1219 01:12:24,310 --> 01:12:27,040 I might ask about. 1220 01:12:27,040 --> 01:12:32,610 I got no particular stake in being arbitrarily evil and 1221 01:12:32,610 --> 01:12:36,520 saying can I find some really teeny little piece of the book 1222 01:12:36,520 --> 01:12:41,910 that I can quiz you about and eh-ha, nobody got it. 1223 01:12:41,910 --> 01:12:44,870 The amusement of that wears off after the first 20 years 1224 01:12:44,870 --> 01:12:47,720 of teaching the course it turns out. 1225 01:12:47,720 --> 01:12:51,970 So, another way of looking at it is I tend to 1226 01:12:51,970 --> 01:12:53,020 ask questions -- 1227 01:12:53,020 --> 01:12:55,250 when I'm going for stuff that's just in the book, I 1228 01:12:55,250 --> 01:12:58,250 tend to go for stuff that's covered in more than one 1229 01:12:58,250 --> 01:12:59,670 sentence in the book. 1230 01:12:59,670 --> 01:13:04,500 So if something is worth a page to Gleitman, you might as 1231 01:13:04,500 --> 01:13:06,870 well pay more attention to that than a 1232 01:13:06,870 --> 01:13:10,930 passing figure caption. 1233 01:13:10,930 --> 01:13:12,440 All right, we'll take one more back there. 1234 01:13:12,440 --> 01:13:14,080 AUDIENCE: Is the exam an hour or an hour and a half? 1235 01:13:14,080 --> 01:13:15,330 PROFESSOR: Hour and a half. 1236 01:13:17,430 --> 01:13:20,660 You should know that it's no longer than it used to be when 1237 01:13:20,660 --> 01:13:22,380 it was an hour long. 1238 01:13:22,380 --> 01:13:23,460 So you'll be fine. 1239 01:13:23,460 --> 01:13:25,480 AUDIENCE: [INAUDIBLE PHRASE]? 1240 01:13:25,480 --> 01:13:27,860 PROFESSOR: What genre of book is it based on? 1241 01:13:27,860 --> 01:13:29,110 Printed. 1242 01:13:34,800 --> 01:13:36,310 Good luck.