0 00:00:00,000 --> 00:00:30,000 Dear viewer, these subtitles were generated by a machine via the service Trint and therefore are (very) buggy. If you are capable, please help us to create good quality subtitles: https://c3subtitles.de/talk/883 Thanks! 1 00:00:15,130 --> 00:00:17,499 The words we use to talk 2 00:00:17,500 --> 00:00:20,109 about a topic strongly influence 3 00:00:20,110 --> 00:00:21,219 how we perceive that 4 00:00:22,330 --> 00:00:23,799 this might seem like an obvious 5 00:00:23,800 --> 00:00:24,800 statement. 6 00:00:25,570 --> 00:00:27,759 But when we look at the news 7 00:00:27,760 --> 00:00:29,919 and what's happening in the last years, 8 00:00:29,920 --> 00:00:32,109 this is something which is so true. 9 00:00:32,110 --> 00:00:34,239 We just think about seriously 10 00:00:34,240 --> 00:00:36,170 a so 11 00:00:37,530 --> 00:00:39,579 this is something we see come true. 12 00:00:39,580 --> 00:00:41,109 Words have a lot of meaning when you 13 00:00:41,110 --> 00:00:42,879 think a Trump what's happening there? 14 00:00:42,880 --> 00:00:45,399 It has a lot to do with speech. 15 00:00:45,400 --> 00:00:47,769 These are it's a linguistics 16 00:00:47,770 --> 00:00:50,139 student and she would talk to us exactly 17 00:00:50,140 --> 00:00:51,549 about this topic. 18 00:00:51,550 --> 00:00:54,009 She would tell us how words can be used 19 00:00:54,010 --> 00:00:55,959 to shape our perception of a certain 20 00:00:55,960 --> 00:00:57,069 topic. 21 00:00:57,070 --> 00:00:59,409 In this case, the 22 00:00:59,410 --> 00:01:02,259 snooper's charter of the UK, 23 00:01:02,260 --> 00:01:04,359 a charter which forces 24 00:01:04,360 --> 00:01:06,909 I speak to save your 25 00:01:06,910 --> 00:01:08,649 browser history. 26 00:01:08,650 --> 00:01:09,909 This is scary. 27 00:01:09,910 --> 00:01:12,009 So please welcome Lisa 28 00:01:12,010 --> 00:01:13,480 was the big round of applause. 29 00:01:17,020 --> 00:01:18,020 Thank you. 30 00:01:21,130 --> 00:01:22,809 OK, so let's begin. 31 00:01:26,500 --> 00:01:29,529 Hi, I'm Lisa, and I'm a linguist. 32 00:01:29,530 --> 00:01:30,879 Welcome to my talk. 33 00:01:30,880 --> 00:01:32,709 This is not a proposal about mass 34 00:01:32,710 --> 00:01:35,469 surveillance, analyzing the terminology 35 00:01:35,470 --> 00:01:37,000 of the U.K. snoopers charter. 36 00:01:38,230 --> 00:01:40,329 So many people interested in linguistics. 37 00:01:40,330 --> 00:01:42,489 So this is amazing. 38 00:01:42,490 --> 00:01:43,989 Thank you very much for attending my 39 00:01:43,990 --> 00:01:44,990 talk. 40 00:01:52,140 --> 00:01:53,879 I hope all of you had a wonderful 41 00:01:53,880 --> 00:01:55,529 Congress so far. 42 00:01:55,530 --> 00:01:57,659 Are there any linguists in the audience? 43 00:01:59,130 --> 00:02:00,130 One 44 00:02:01,800 --> 00:02:04,019 and maybe some more yet 45 00:02:04,020 --> 00:02:04,919 in the back. 46 00:02:04,920 --> 00:02:06,419 Nice. Very nice. 47 00:02:06,420 --> 00:02:08,729 OK, um, let's begin, 48 00:02:09,810 --> 00:02:12,089 um, in this talk, 49 00:02:12,090 --> 00:02:14,159 I'm going to take a closer look at the 50 00:02:14,160 --> 00:02:16,619 Snooper's Charter and its terminology 51 00:02:16,620 --> 00:02:19,109 from a corpus linguistic point of view. 52 00:02:19,110 --> 00:02:21,749 So what is 53 00:02:21,750 --> 00:02:24,029 the Investigatory Powers Act, also 54 00:02:24,030 --> 00:02:25,030 known as the charter? 55 00:02:26,100 --> 00:02:28,859 It is a new legislation which extends 56 00:02:28,860 --> 00:02:30,509 the surveillance powers of the state, 57 00:02:31,800 --> 00:02:34,559 and it would allow the security agencies 58 00:02:34,560 --> 00:02:36,449 to keep so-called Internet connection 59 00:02:36,450 --> 00:02:39,089 records for up to one year 60 00:02:39,090 --> 00:02:41,789 and to hack mobile phones and laptops 61 00:02:41,790 --> 00:02:44,069 on a large scale without any suspicion. 62 00:02:45,360 --> 00:02:47,459 It was previously 63 00:02:47,460 --> 00:02:49,739 known as the Investigatory Powers 64 00:02:49,740 --> 00:02:51,029 Bill. 65 00:02:51,030 --> 00:02:52,559 And to make it a bit clearer. 66 00:02:52,560 --> 00:02:54,779 A bill is a law 67 00:02:54,780 --> 00:02:56,879 which has only been proposed but 68 00:02:56,880 --> 00:02:58,979 has not passed both 69 00:02:58,980 --> 00:03:00,869 houses of Parliament. 70 00:03:00,870 --> 00:03:02,969 The Investigatory Powers Bill 71 00:03:02,970 --> 00:03:05,069 has received royal assent 72 00:03:05,070 --> 00:03:07,289 and was made into an act in 73 00:03:07,290 --> 00:03:09,020 November 2016. 74 00:03:10,200 --> 00:03:12,509 It was nicknamed Snooper's Charter 75 00:03:12,510 --> 00:03:14,639 due to the extensive surveillance 76 00:03:16,380 --> 00:03:18,359 to give a bit of history. 77 00:03:18,360 --> 00:03:20,609 First term Snooper's Charter has been 78 00:03:20,610 --> 00:03:23,099 around for a while and 79 00:03:23,100 --> 00:03:24,809 it is not new at all. 80 00:03:24,810 --> 00:03:27,119 It was used for the communications 81 00:03:27,120 --> 00:03:29,369 data bill. Previously, a 82 00:03:29,370 --> 00:03:32,159 previous surveillance bill 83 00:03:32,160 --> 00:03:34,409 and a newspaper article 84 00:03:34,410 --> 00:03:36,869 refers to this bill as 85 00:03:36,870 --> 00:03:39,359 the most draconian piece of surveillance 86 00:03:39,360 --> 00:03:41,429 legislation ever proposed in a 87 00:03:41,430 --> 00:03:42,359 democracy. 88 00:03:42,360 --> 00:03:43,739 Wow. 89 00:03:43,740 --> 00:03:46,169 So this bill was blocked 90 00:03:46,170 --> 00:03:47,550 in 2013, 91 00:03:48,600 --> 00:03:50,819 but a new bill, as we can see, is 92 00:03:50,820 --> 00:03:52,189 on the way. 93 00:03:52,190 --> 00:03:55,049 So, um, 94 00:03:55,050 --> 00:03:57,089 now I'm going to talk briefly about the 95 00:03:57,090 --> 00:03:59,669 methods I used to give you 96 00:03:59,670 --> 00:04:02,369 an impression of what I was doing. 97 00:04:02,370 --> 00:04:04,439 Um, as I want to find 98 00:04:04,440 --> 00:04:06,579 out how certain expressions are used in 99 00:04:06,580 --> 00:04:08,789 the parliamentary debates and in 100 00:04:08,790 --> 00:04:11,459 the media coverage, um, 101 00:04:11,460 --> 00:04:13,529 my focus will be on the context 102 00:04:13,530 --> 00:04:16,259 of certain words and words, 103 00:04:16,260 --> 00:04:18,539 as you know, do not stand on its own, 104 00:04:18,540 --> 00:04:20,639 but always have a context around it 105 00:04:20,640 --> 00:04:22,709 to explain the meaning and to 106 00:04:22,710 --> 00:04:24,089 give an impression of what the word is 107 00:04:24,090 --> 00:04:25,090 actually about. 108 00:04:27,270 --> 00:04:29,969 OK, so 109 00:04:29,970 --> 00:04:32,189 I'm going to make use of methods 110 00:04:32,190 --> 00:04:33,869 of corpus linguistics. 111 00:04:33,870 --> 00:04:35,340 What is corpus linguistics? 112 00:04:36,510 --> 00:04:38,669 Linguistics is the study of language 113 00:04:38,670 --> 00:04:40,739 based on examples of real life 114 00:04:40,740 --> 00:04:41,740 language use. 115 00:04:42,960 --> 00:04:44,819 What is real life language? 116 00:04:44,820 --> 00:04:47,689 Well, newspaper articles, leaflets, 117 00:04:47,690 --> 00:04:50,009 student essays, transcripts 118 00:04:50,010 --> 00:04:51,689 of spoken data and so on. 119 00:04:53,250 --> 00:04:55,619 Corpus is a collection of machine 120 00:04:55,620 --> 00:04:58,199 readable texts and corpora 121 00:04:58,200 --> 00:05:00,509 can come in various sizes 122 00:05:00,510 --> 00:05:03,059 ranging from several thousand 123 00:05:03,060 --> 00:05:05,129 to millions, even billions of 124 00:05:05,130 --> 00:05:06,479 words. 125 00:05:06,480 --> 00:05:07,480 To give an example, 126 00:05:08,880 --> 00:05:11,069 Wikipedia, everybody knows it 127 00:05:11,070 --> 00:05:13,799 can also be regarded as a corpus 128 00:05:13,800 --> 00:05:16,109 and contains around one point 129 00:05:16,110 --> 00:05:17,399 nine billion words. 130 00:05:17,400 --> 00:05:20,609 It's really a large corpus, 131 00:05:20,610 --> 00:05:22,919 whereas the British National Corpus 132 00:05:22,920 --> 00:05:25,259 is a bit smaller and contains 133 00:05:25,260 --> 00:05:27,629 around 100 million words. 134 00:05:27,630 --> 00:05:29,789 So why 135 00:05:29,790 --> 00:05:32,159 is corpus linguistics useful? 136 00:05:32,160 --> 00:05:34,979 Well, it allows you to work with 137 00:05:34,980 --> 00:05:37,259 large amounts of data and 138 00:05:37,260 --> 00:05:39,359 linguistics. Examinations have 139 00:05:39,360 --> 00:05:41,219 been carried out using software, for 140 00:05:41,220 --> 00:05:44,249 example. And in my case, 141 00:05:44,250 --> 00:05:47,369 and it allows you to work with 142 00:05:47,370 --> 00:05:49,469 amounts of data which which would 143 00:05:49,470 --> 00:05:51,509 take years to analyze by hand, and you 144 00:05:51,510 --> 00:05:52,889 can just do it with 145 00:05:54,480 --> 00:05:55,480 a lot easier. 146 00:05:57,840 --> 00:05:58,840 Um, 147 00:06:01,800 --> 00:06:03,179 why is it also useful? 148 00:06:03,180 --> 00:06:05,519 Well, it helps to reduce research, 149 00:06:05,520 --> 00:06:08,069 a bias that means that 150 00:06:08,070 --> 00:06:10,139 there is a tendency the researchers 151 00:06:10,140 --> 00:06:12,359 tend to pay more attention to things 152 00:06:12,360 --> 00:06:14,309 they encounter at an earlier stage of 153 00:06:14,310 --> 00:06:16,859 research than at a later stage. 154 00:06:16,860 --> 00:06:18,959 And if you're working with 155 00:06:18,960 --> 00:06:21,059 millions of words, it's a lot harder 156 00:06:21,060 --> 00:06:24,299 to be biased about your results 157 00:06:24,300 --> 00:06:26,459 compared to only work with, let's 158 00:06:26,460 --> 00:06:28,469 say, two or three newspaper articles 159 00:06:28,470 --> 00:06:29,910 which are analyzed by hand. 160 00:06:32,040 --> 00:06:33,040 So. 161 00:06:34,440 --> 00:06:36,809 OK, this analysis I used 162 00:06:36,810 --> 00:06:39,539 to methods of linguistics, 163 00:06:39,540 --> 00:06:41,629 concordance analysis and 164 00:06:41,630 --> 00:06:43,769 colocation analysis, 165 00:06:43,770 --> 00:06:45,839 both methods of very 166 00:06:45,840 --> 00:06:47,939 useful to examine the context of 167 00:06:47,940 --> 00:06:48,959 keywords in the corpus. 168 00:06:52,120 --> 00:06:54,669 OK, concordance analysis 169 00:06:54,670 --> 00:06:57,219 gives you several words around 170 00:06:57,220 --> 00:06:59,409 the key word you were searching for, 171 00:06:59,410 --> 00:07:01,479 and the span of words can be adjusted 172 00:07:01,480 --> 00:07:03,669 and usually it's around five words 173 00:07:03,670 --> 00:07:05,019 to either side of the keyboard. 174 00:07:06,430 --> 00:07:09,009 So this is a screenshot from 175 00:07:09,010 --> 00:07:11,769 the corpus analysis program I used 176 00:07:11,770 --> 00:07:13,689 in the middle and low script. 177 00:07:13,690 --> 00:07:15,759 You can see bug data. 178 00:07:15,760 --> 00:07:17,859 This is my keyboard, which I ended at 179 00:07:17,860 --> 00:07:20,109 the very bottom, if you can see. 180 00:07:20,110 --> 00:07:22,449 And uncuffing 181 00:07:22,450 --> 00:07:23,770 allows you to 182 00:07:25,030 --> 00:07:26,859 analyze three levels. 183 00:07:26,860 --> 00:07:29,139 Four is, in this case, one word to 184 00:07:29,140 --> 00:07:31,149 the left of the keyword is the red 185 00:07:31,150 --> 00:07:33,489 section and 186 00:07:33,490 --> 00:07:35,589 the second and the third 187 00:07:35,590 --> 00:07:38,649 word to the right of the keyboard. 188 00:07:38,650 --> 00:07:41,259 You can see in green and 189 00:07:41,260 --> 00:07:42,199 purple. 190 00:07:42,200 --> 00:07:44,259 And in this version, 191 00:07:44,260 --> 00:07:45,949 it's sorted alphabetically. 192 00:07:45,950 --> 00:07:48,099 Then you can scroll down and have a look 193 00:07:48,100 --> 00:07:50,169 at the other concrete answers and 194 00:07:50,170 --> 00:07:52,449 all the contexts around type data. 195 00:07:55,370 --> 00:07:57,559 It's the second method I used 196 00:07:57,560 --> 00:08:00,319 is called colocation analysis, 197 00:08:00,320 --> 00:08:02,569 and this method is very useful 198 00:08:02,570 --> 00:08:04,819 to examine the connotations and 199 00:08:04,820 --> 00:08:06,410 associations between words 200 00:08:07,790 --> 00:08:09,949 of the relation to other words is crucial 201 00:08:09,950 --> 00:08:11,759 for understanding the meaning of a word 202 00:08:11,760 --> 00:08:14,029 we already had. That and 203 00:08:14,030 --> 00:08:16,189 the location is defined as 204 00:08:16,190 --> 00:08:18,349 the above frequent occurrence of 205 00:08:18,350 --> 00:08:20,689 two words within a predetermined 206 00:08:20,690 --> 00:08:23,389 span. Once again, it's usually 207 00:08:23,390 --> 00:08:25,579 around five words to either side 208 00:08:25,580 --> 00:08:28,219 of the search term, because if you 209 00:08:28,220 --> 00:08:30,379 make the span larger, you might 210 00:08:30,380 --> 00:08:32,569 have two or three, 211 00:08:32,570 --> 00:08:34,849 two or three sentences 212 00:08:34,850 --> 00:08:36,739 which are not even related to a key word 213 00:08:36,740 --> 00:08:37,740 anymore. 214 00:08:38,350 --> 00:08:39,350 Hmm. 215 00:08:39,740 --> 00:08:41,808 And to give an example, 216 00:08:41,809 --> 00:08:43,908 the two strongest collocations and 217 00:08:43,909 --> 00:08:46,249 the British National Corpus for Juice 218 00:08:46,250 --> 00:08:48,139 are orange and lemon. 219 00:08:48,140 --> 00:08:50,429 So, OK. 220 00:08:52,070 --> 00:08:54,379 And in order to indicate 221 00:08:54,380 --> 00:08:56,839 the strength of a colocation mutual 222 00:08:56,840 --> 00:08:58,999 information is used, it's 223 00:08:59,000 --> 00:09:01,189 calculated based on 224 00:09:01,190 --> 00:09:03,259 the expected probability of two 225 00:09:03,260 --> 00:09:05,899 words appearing in each other, 226 00:09:05,900 --> 00:09:08,479 their relative frequencies and 227 00:09:08,480 --> 00:09:11,119 the overall size of a corpus. 228 00:09:11,120 --> 00:09:13,219 Then you compare the 229 00:09:13,220 --> 00:09:15,589 expected probability and the observed 230 00:09:15,590 --> 00:09:17,779 probability, and this difference 231 00:09:17,780 --> 00:09:19,969 is translated into a number. 232 00:09:19,970 --> 00:09:21,919 And the higher the number, the stronger 233 00:09:21,920 --> 00:09:23,359 the location. 234 00:09:23,360 --> 00:09:25,519 But if you use a 235 00:09:25,520 --> 00:09:27,619 corpus analysis program, all this 236 00:09:27,620 --> 00:09:28,820 is done for you. So 237 00:09:29,930 --> 00:09:32,539 here's a screenshot and 238 00:09:32,540 --> 00:09:35,329 you can see the Concord collocations 239 00:09:35,330 --> 00:09:36,330 of surveillance. 240 00:09:38,210 --> 00:09:40,459 And this is quite unusual 241 00:09:40,460 --> 00:09:43,489 that so many birds have the same value, 242 00:09:43,490 --> 00:09:45,079 same value. 243 00:09:45,080 --> 00:09:47,479 You can see it in the section with 244 00:09:47,480 --> 00:09:48,529 such statistics. 245 00:09:56,230 --> 00:09:58,929 And usually articles, 246 00:09:58,930 --> 00:10:00,999 prepositions cetera, are not 247 00:10:01,000 --> 00:10:03,429 taken into consideration 248 00:10:03,430 --> 00:10:06,399 because the frequency is 249 00:10:06,400 --> 00:10:08,679 evenly distributed across the corpus. 250 00:10:10,060 --> 00:10:12,009 OK, so 251 00:10:13,240 --> 00:10:15,519 in my analysis, I used to 252 00:10:15,520 --> 00:10:17,709 cover up the Investigatory 253 00:10:17,710 --> 00:10:19,839 Powers Act corpus and the news 254 00:10:19,840 --> 00:10:20,840 on the Web corpus. 255 00:10:23,380 --> 00:10:26,859 So the Investigatory Powers Act corpus 256 00:10:26,860 --> 00:10:29,769 was compiled by searching for 257 00:10:29,770 --> 00:10:31,959 the expression investigatory powers 258 00:10:31,960 --> 00:10:34,299 bill on Hans-Gert, the 259 00:10:34,300 --> 00:10:36,609 official documentation of parliamentary 260 00:10:36,610 --> 00:10:38,679 debates and my 261 00:10:38,680 --> 00:10:40,929 corpus consists of sixty three 262 00:10:40,930 --> 00:10:43,569 debates, some written statements 263 00:10:43,570 --> 00:10:45,699 and the final legal text of the 264 00:10:45,700 --> 00:10:48,549 Investigatory Powers Act until 265 00:10:48,550 --> 00:10:51,099 it has a size of one million 266 00:10:51,100 --> 00:10:53,329 sixty thousand two hundred and nineteen 267 00:10:53,330 --> 00:10:54,429 words. 268 00:10:54,430 --> 00:10:57,249 So it's a relatively small corpus, 269 00:10:57,250 --> 00:11:00,219 but still it's like it's large enough to 270 00:11:00,220 --> 00:11:02,289 show some patterns 271 00:11:02,290 --> 00:11:04,359 and it's not too small to not 272 00:11:04,360 --> 00:11:05,690 be able to do an analysis. 273 00:11:07,900 --> 00:11:11,019 The second corpus is an online corpus 274 00:11:11,020 --> 00:11:13,419 called Nese on the web, 275 00:11:13,420 --> 00:11:15,819 and it consists of newspaper 276 00:11:15,820 --> 00:11:19,119 articles from the English speaking world. 277 00:11:19,120 --> 00:11:21,249 And it was 278 00:11:21,250 --> 00:11:23,379 started in 2010 and it 279 00:11:23,380 --> 00:11:25,749 is updated every 24 hours. 280 00:11:25,750 --> 00:11:28,209 So it is the most recent 281 00:11:28,210 --> 00:11:30,279 corpus you can find because it covers 282 00:11:30,280 --> 00:11:32,740 all the articles until yesterday. 283 00:11:35,590 --> 00:11:37,749 And in order to 284 00:11:37,750 --> 00:11:40,539 be able to compare the two corpora 285 00:11:40,540 --> 00:11:43,089 I restrict restricted the time span 286 00:11:43,090 --> 00:11:45,549 of the searches to the time span 287 00:11:45,550 --> 00:11:47,619 of the parliamentary debate, that 288 00:11:47,620 --> 00:11:50,799 is from July 2014 289 00:11:50,800 --> 00:11:53,110 until November 2016. 290 00:11:54,640 --> 00:11:56,979 All in all, in this timespan 291 00:11:56,980 --> 00:11:59,199 I was interested in the corpus 292 00:11:59,200 --> 00:12:01,539 had four billion, 293 00:12:01,540 --> 00:12:03,949 350 million words. 294 00:12:03,950 --> 00:12:06,249 So it is roughly four thousand 295 00:12:06,250 --> 00:12:08,979 times larger than the investigatory 296 00:12:08,980 --> 00:12:09,980 powers that corpus. 297 00:12:11,500 --> 00:12:13,839 So with such a vast amount 298 00:12:13,840 --> 00:12:16,029 of data, you, of course, can do a lot 299 00:12:16,030 --> 00:12:17,889 of different analysis. 300 00:12:17,890 --> 00:12:20,079 And this is how I will pick out three 301 00:12:20,080 --> 00:12:21,080 keywords. 302 00:12:21,730 --> 00:12:23,829 And this is going to be the term 303 00:12:23,830 --> 00:12:26,349 snooper's charter mass surveillance, 304 00:12:26,350 --> 00:12:29,319 everyone's favorite word and thought, 305 00:12:29,320 --> 00:12:32,289 OK, let's begin. 306 00:12:32,290 --> 00:12:34,899 So in order to find out what 307 00:12:34,900 --> 00:12:38,049 to Snoop means, I consult the dictionary 308 00:12:38,050 --> 00:12:40,869 and the Oxford English Dictionary 309 00:12:40,870 --> 00:12:43,029 to Snoop is to find us 310 00:12:43,030 --> 00:12:45,249 to pry into matters one need not be 311 00:12:45,250 --> 00:12:46,250 concerned with. 312 00:12:47,800 --> 00:12:50,319 Keep that in mind for that list. 313 00:12:50,320 --> 00:12:52,929 So let's have a look at the concurrences. 314 00:12:53,980 --> 00:12:55,809 Usually, as you've seen on the 315 00:12:55,810 --> 00:12:57,879 screenshot, you have 50 316 00:12:57,880 --> 00:12:59,649 or even 100 concrete answers. 317 00:12:59,650 --> 00:13:01,989 But in this talk, I will only give you 318 00:13:01,990 --> 00:13:04,359 a small excerpt, 319 00:13:04,360 --> 00:13:05,360 so. 320 00:13:07,160 --> 00:13:09,439 In my view, it will knock out completely 321 00:13:09,440 --> 00:13:11,520 that lazy label of snooper's charter. 322 00:13:13,460 --> 00:13:15,679 I was not going to use the phrase stupid 323 00:13:15,680 --> 00:13:18,559 charter because it is counterproductive, 324 00:13:18,560 --> 00:13:20,779 nor was I going to use the phrase mass 325 00:13:20,780 --> 00:13:21,780 surveillance. 326 00:13:22,760 --> 00:13:24,799 I was about to attack him for the phrase 327 00:13:24,800 --> 00:13:26,869 snooper's charter, but 328 00:13:26,870 --> 00:13:28,370 he managed to get out in time. 329 00:13:29,690 --> 00:13:31,939 I also hope that the politically loaded 330 00:13:31,940 --> 00:13:34,129 and seriously misleading phrase 331 00:13:34,130 --> 00:13:36,289 Snooper's charter has been removed 332 00:13:36,290 --> 00:13:39,289 from our lexicon snooping, 333 00:13:39,290 --> 00:13:41,389 hence the populist phrase snooper's 334 00:13:41,390 --> 00:13:42,390 charter. 335 00:13:43,010 --> 00:13:46,009 So what is striking about 336 00:13:46,010 --> 00:13:48,079 the expression Snooper's Charter 337 00:13:48,080 --> 00:13:50,269 is that it's either accompanied 338 00:13:50,270 --> 00:13:52,429 by a direct form of direct 339 00:13:52,430 --> 00:13:54,859 negation or another form 340 00:13:54,860 --> 00:13:57,259 of denunciation or disapproval, 341 00:13:57,260 --> 00:13:59,629 such as lazy label, 342 00:13:59,630 --> 00:14:01,879 misleading phrase, populars 343 00:14:01,880 --> 00:14:05,269 phrase, conspiracy theories 344 00:14:05,270 --> 00:14:06,529 and so on and so forth. 345 00:14:07,640 --> 00:14:09,769 And all this aims at 346 00:14:09,770 --> 00:14:11,959 invalidating criticism of the bill. 347 00:14:14,300 --> 00:14:16,369 In addition, it conveys the 348 00:14:16,370 --> 00:14:18,469 message that everybody who refers 349 00:14:18,470 --> 00:14:20,899 to the bill as a snooper's charter 350 00:14:20,900 --> 00:14:23,239 is neither trustworthy nor 351 00:14:23,240 --> 00:14:25,309 informed, nor appreciates the 352 00:14:25,310 --> 00:14:27,529 work of security agencies 353 00:14:27,530 --> 00:14:29,649 and have a wonderful quote for you. 354 00:14:31,730 --> 00:14:33,799 In my view, it is lazy to label the 355 00:14:33,800 --> 00:14:36,109 bill a snoopers charter or a plan 356 00:14:36,110 --> 00:14:37,519 for mass surveillance. 357 00:14:37,520 --> 00:14:40,369 In fact, it is worse than lazy. 358 00:14:40,370 --> 00:14:42,409 It is insulting to people who work at the 359 00:14:42,410 --> 00:14:44,840 police and security services. 360 00:14:46,610 --> 00:14:48,949 So thank you, Bill. 361 00:14:48,950 --> 00:14:51,349 And the term to Snoop is highly 362 00:14:51,350 --> 00:14:52,639 undesired. 363 00:14:52,640 --> 00:14:54,709 Remember the definition? 364 00:14:54,710 --> 00:14:56,509 It implies that the government 365 00:14:56,510 --> 00:14:58,669 overstepped its bounds and that 366 00:14:58,670 --> 00:15:00,799 the bill legalizes 367 00:15:00,800 --> 00:15:02,569 activities which are otherwise 368 00:15:02,570 --> 00:15:03,570 unacceptable. 369 00:15:05,510 --> 00:15:07,819 The strongest case of 370 00:15:07,820 --> 00:15:10,519 snooper's charter, a label, 371 00:15:10,520 --> 00:15:12,469 phrase and court. 372 00:15:12,470 --> 00:15:14,689 And all this suggests that 373 00:15:14,690 --> 00:15:17,149 a snooper's charter and the genuine 374 00:15:17,150 --> 00:15:19,369 nature of the bill are two 375 00:15:19,370 --> 00:15:21,529 entirely different things. 376 00:15:21,530 --> 00:15:23,839 So Snooper's Charter is 377 00:15:23,840 --> 00:15:26,149 nothing but a hollow phrase, 378 00:15:26,150 --> 00:15:28,039 and it lacks an element. 379 00:15:28,040 --> 00:15:29,040 It signifies. 380 00:15:30,130 --> 00:15:31,459 OK, so 381 00:15:32,510 --> 00:15:35,149 let's have a look at how Superchargers 382 00:15:35,150 --> 00:15:36,649 use in the newspaper. 383 00:15:36,650 --> 00:15:37,650 Kobes 384 00:15:39,380 --> 00:15:41,899 reintroduce a beefed up version 385 00:15:41,900 --> 00:15:44,119 of the Snooper's Charter, 386 00:15:44,120 --> 00:15:46,879 the superchargers discredited 387 00:15:46,880 --> 00:15:49,099 unlawful and treats us all 388 00:15:49,100 --> 00:15:50,100 as suspects, 389 00:15:51,200 --> 00:15:53,029 saying that the bill is neither a 390 00:15:53,030 --> 00:15:55,159 Snooper's charter nor a plan for 391 00:15:55,160 --> 00:15:57,589 mass surveillance. A renewed 392 00:15:57,590 --> 00:16:00,139 effort to pass the snoopers charter bill 393 00:16:00,140 --> 00:16:01,940 of increased surveillance powers. 394 00:16:03,230 --> 00:16:05,489 The Home Secretary will try to sweeten 395 00:16:05,490 --> 00:16:07,999 the pill of Revived Stupas 396 00:16:08,000 --> 00:16:09,000 Charter. 397 00:16:09,920 --> 00:16:12,979 OK, so roughly 398 00:16:12,980 --> 00:16:15,049 one third of 399 00:16:15,050 --> 00:16:17,119 the concrete answers mentioned 400 00:16:17,120 --> 00:16:19,459 that the investigatory powers bill 401 00:16:19,460 --> 00:16:22,159 is named Snooper's Charter, 402 00:16:22,160 --> 00:16:24,529 and more than one third mentions 403 00:16:24,530 --> 00:16:26,299 the existence of a previous snoopers 404 00:16:26,300 --> 00:16:28,429 charter, the communications data 405 00:16:28,430 --> 00:16:30,679 bill. And to be honest, this 406 00:16:30,680 --> 00:16:32,959 is how I knew about 407 00:16:32,960 --> 00:16:35,059 this communications data bill, because I 408 00:16:35,060 --> 00:16:37,279 did some research only 409 00:16:37,280 --> 00:16:39,349 due to reading about it on 410 00:16:39,350 --> 00:16:42,019 a newspaper, not by doing research 411 00:16:42,020 --> 00:16:43,309 on parliamentary debates. 412 00:16:43,310 --> 00:16:44,809 Corbetts So 413 00:16:47,090 --> 00:16:49,869 this is why newspapers report about 414 00:16:49,870 --> 00:16:52,159 renewed effort to reintroduce 415 00:16:52,160 --> 00:16:53,899 a beefed up version of the snoopers 416 00:16:53,900 --> 00:16:54,900 charter. 417 00:16:55,480 --> 00:16:58,219 So as a consequence, 418 00:16:58,220 --> 00:17:00,099 the negative overturn of the previous 419 00:17:00,100 --> 00:17:02,109 communications data bill is transferred 420 00:17:02,110 --> 00:17:03,110 to the current bill. 421 00:17:05,510 --> 00:17:07,549 In addition, it is pointed out that the 422 00:17:07,550 --> 00:17:09,858 bill is made to seem less unpleasant 423 00:17:09,859 --> 00:17:11,959 than it actually is, for the 424 00:17:11,960 --> 00:17:14,118 home secretary will try to sweeten 425 00:17:14,119 --> 00:17:15,858 the pill of her revived snooper's 426 00:17:15,859 --> 00:17:16,969 charter. 427 00:17:16,970 --> 00:17:19,249 So it's 428 00:17:19,250 --> 00:17:21,618 not a sad reality occurs 429 00:17:21,619 --> 00:17:23,328 with the allegation and the newspaper 430 00:17:23,329 --> 00:17:24,259 articles. 431 00:17:24,260 --> 00:17:26,088 And if it does occur within the 432 00:17:26,089 --> 00:17:28,219 negotiation, it is a quote from 433 00:17:28,220 --> 00:17:29,329 the parliamentary debate. 434 00:17:31,070 --> 00:17:34,039 And the strongest colocated of 435 00:17:34,040 --> 00:17:36,169 snooper's charter are so-called 436 00:17:36,170 --> 00:17:37,170 and. 437 00:17:37,790 --> 00:17:39,980 OK, so. 438 00:17:46,020 --> 00:17:47,729 What can we learn from the concrete 439 00:17:47,730 --> 00:17:49,140 answers about this expression? 440 00:17:51,950 --> 00:17:54,289 So in the investigatory powers, 441 00:17:54,290 --> 00:17:57,409 that cop is 70 percent, 442 00:17:57,410 --> 00:18:00,349 75 percent of the occurrences 443 00:18:00,350 --> 00:18:02,749 are negative or contain some 444 00:18:02,750 --> 00:18:05,119 form of disapproval or denunciation 445 00:18:06,160 --> 00:18:06,879 of terrorism. 446 00:18:06,880 --> 00:18:08,359 Now, Corpus. 447 00:18:08,360 --> 00:18:10,669 The term is hardly negated. 448 00:18:10,670 --> 00:18:12,649 The Investigatory Powers Bill entirely 449 00:18:12,650 --> 00:18:16,129 lacks reference to previous supercharger. 450 00:18:16,130 --> 00:18:18,679 But the newspaper Corpus, many references 451 00:18:18,680 --> 00:18:20,409 are given to the previous judge. 452 00:18:22,860 --> 00:18:24,779 And the parliamentary debates criticism 453 00:18:24,780 --> 00:18:26,879 is used to refer to 454 00:18:26,880 --> 00:18:29,759 people using the term 455 00:18:29,760 --> 00:18:32,669 where it's in the newspaper corpus 456 00:18:32,670 --> 00:18:35,309 criticism is directed at the implications 457 00:18:35,310 --> 00:18:36,310 of the bill. 458 00:18:37,590 --> 00:18:39,739 So all in all, the 459 00:18:39,740 --> 00:18:41,789 snooper's charge is burdened with 460 00:18:41,790 --> 00:18:43,949 negative associations from the previous 461 00:18:43,950 --> 00:18:46,019 communications data bill 462 00:18:46,020 --> 00:18:48,269 and as a consequence, it is strictly 463 00:18:48,270 --> 00:18:50,489 avoided in the parliamentary debates and 464 00:18:50,490 --> 00:18:53,219 if it is used at all, it is negated. 465 00:18:53,220 --> 00:18:54,220 OK, so 466 00:18:55,500 --> 00:18:58,079 second term, I had a look at this 467 00:18:58,080 --> 00:18:59,080 mass surveillance. 468 00:19:02,180 --> 00:19:04,369 It is not important, is it not important 469 00:19:04,370 --> 00:19:07,189 for us to continue to reassure the public 470 00:19:07,190 --> 00:19:09,199 that this is not a proposal for mass 471 00:19:09,200 --> 00:19:11,329 surveillance and to restate the 472 00:19:11,330 --> 00:19:12,559 essential need for the bill? 473 00:19:20,880 --> 00:19:23,039 Not accept that the bill is a plan for 474 00:19:23,040 --> 00:19:24,569 mass surveillance. 475 00:19:24,570 --> 00:19:26,819 Neither is snooper's charter nor a plan 476 00:19:26,820 --> 00:19:28,949 for mass surveillance, 477 00:19:28,950 --> 00:19:31,049 a snoopers charter or a plan for 478 00:19:31,050 --> 00:19:32,699 mass surveillance. 479 00:19:32,700 --> 00:19:34,799 Neither is stupid charter nor a plan for 480 00:19:34,800 --> 00:19:36,209 mass surveillance. 481 00:19:36,210 --> 00:19:37,259 So altogether, 482 00:19:38,610 --> 00:19:41,069 the term mass surveillance 483 00:19:41,070 --> 00:19:43,229 does not appear very often throughout the 484 00:19:43,230 --> 00:19:45,539 entire corpus, about 485 00:19:45,540 --> 00:19:47,579 28 times per million words. 486 00:19:47,580 --> 00:19:48,900 Is not a lot. 487 00:19:50,190 --> 00:19:51,929 And in over 50 percent of these 488 00:19:51,930 --> 00:19:54,390 occurrences, a firm of negation appears 489 00:19:55,440 --> 00:19:57,719 in several instances mentioned 490 00:19:57,720 --> 00:19:59,729 snooper's charter and mass surveillance 491 00:19:59,730 --> 00:20:01,799 in one negative sentence in 492 00:20:01,800 --> 00:20:03,929 order to suggest that both terms 493 00:20:03,930 --> 00:20:06,149 are equally inadequate to describe the 494 00:20:06,150 --> 00:20:07,619 Investigatory Powers Bill. 495 00:20:09,430 --> 00:20:11,729 Strong Colgate's of 496 00:20:11,730 --> 00:20:14,099 mass surveillance plan 497 00:20:14,100 --> 00:20:15,059 Chaja. 498 00:20:15,060 --> 00:20:16,589 And of course, not 499 00:20:18,540 --> 00:20:19,540 OK. 500 00:20:21,510 --> 00:20:23,639 So and 501 00:20:23,640 --> 00:20:25,979 here's some excerpts from 502 00:20:25,980 --> 00:20:27,630 the news on the Web, corpus 503 00:20:29,940 --> 00:20:32,259 open court hearing into whether 504 00:20:32,260 --> 00:20:34,559 J'accuse alleged mass surveillance 505 00:20:34,560 --> 00:20:36,629 of private communication breaks, 506 00:20:36,630 --> 00:20:38,939 human right laws, major U.S. 507 00:20:38,940 --> 00:20:41,039 technology companies suffering from the 508 00:20:41,040 --> 00:20:43,139 fallout of the NSA 509 00:20:43,140 --> 00:20:45,629 mass surveillance programs 510 00:20:45,630 --> 00:20:47,999 under criticism for its mass surveillance 511 00:20:48,000 --> 00:20:49,589 practices. 512 00:20:49,590 --> 00:20:51,899 Reports about the NSA mass 513 00:20:51,900 --> 00:20:52,949 surveillance programs 514 00:20:54,150 --> 00:20:56,249 carrying out mass surveillance as their 515 00:20:56,250 --> 00:20:57,989 critics have claimed. 516 00:20:57,990 --> 00:21:00,119 So the concrete answers 517 00:21:00,120 --> 00:21:02,309 reveal a strong negative connotation 518 00:21:02,310 --> 00:21:04,169 of mass surveillance. 519 00:21:04,170 --> 00:21:06,419 It ranges from breaking human 520 00:21:06,420 --> 00:21:08,879 right laws to 521 00:21:08,880 --> 00:21:10,349 even the nuclear fallout. 522 00:21:12,410 --> 00:21:14,839 And by drawing these comparisons, 523 00:21:14,840 --> 00:21:16,819 it is implied that the consequences are 524 00:21:16,820 --> 00:21:18,979 harmful, long lasting 525 00:21:18,980 --> 00:21:20,959 and not easily to overcome. 526 00:21:20,960 --> 00:21:23,240 So mass surveillance is dangerous. 527 00:21:24,260 --> 00:21:26,809 This is a message, and 528 00:21:26,810 --> 00:21:29,029 this was only an excerpt, but other 529 00:21:29,030 --> 00:21:31,159 concrete answers mention mass 530 00:21:31,160 --> 00:21:33,109 surveillance in the same breath as war 531 00:21:33,110 --> 00:21:35,419 crimes and totalitarian 532 00:21:35,420 --> 00:21:37,939 institutions such as the Nazis 533 00:21:37,940 --> 00:21:39,609 and the East German secret police. 534 00:21:40,850 --> 00:21:43,009 Some articles refer directly to 535 00:21:43,010 --> 00:21:45,289 mass surveillance as bulk 536 00:21:45,290 --> 00:21:48,229 collection, and I give you a quote, 537 00:21:48,230 --> 00:21:50,209 weaseling their way out by calling mass 538 00:21:50,210 --> 00:21:52,549 surveillance terms like bulk collection. 539 00:21:54,230 --> 00:21:55,999 Very few instances bring up mass 540 00:21:56,000 --> 00:21:58,159 surveillance in a negative sentence 541 00:21:58,160 --> 00:22:00,319 that most of these instances are quotes 542 00:22:00,320 --> 00:22:02,449 from politicians or even from the 543 00:22:02,450 --> 00:22:04,369 debates regarding the investigatory 544 00:22:04,370 --> 00:22:05,959 powers bill. 545 00:22:05,960 --> 00:22:08,089 Strong as a case of 546 00:22:08,090 --> 00:22:11,049 mass surveillance, Tempora 547 00:22:11,050 --> 00:22:13,339 GSB, which is which stands 548 00:22:13,340 --> 00:22:14,959 for New Zealand's Government 549 00:22:14,960 --> 00:22:17,269 Communications Security Bureau 550 00:22:17,270 --> 00:22:19,279 and GCE HQ. 551 00:22:19,280 --> 00:22:21,379 So mass surveillance is 552 00:22:21,380 --> 00:22:23,749 inevitably linked to modern governments 553 00:22:23,750 --> 00:22:25,279 and the surveillance programs. 554 00:22:27,260 --> 00:22:29,779 And if you omit acronyms 555 00:22:29,780 --> 00:22:31,549 because they do not really count as 556 00:22:31,550 --> 00:22:33,679 lexical, um, the 557 00:22:33,680 --> 00:22:35,839 most the strongest 558 00:22:35,840 --> 00:22:39,079 collocated are revelations, 559 00:22:39,080 --> 00:22:40,980 indiscriminate and interception. 560 00:22:41,990 --> 00:22:42,990 So 561 00:22:44,420 --> 00:22:46,839 what can we learn from the conferences? 562 00:22:48,080 --> 00:22:50,359 Well, in the parliamentary debates 563 00:22:50,360 --> 00:22:52,369 and over 50 percent of the occurrences, 564 00:22:52,370 --> 00:22:54,619 mass surveillance is negated 565 00:22:54,620 --> 00:22:56,779 and now it is hardly negative. 566 00:22:56,780 --> 00:22:59,389 And now Corpuz also reveals 567 00:22:59,390 --> 00:23:01,999 the full extent of mass surveillance 568 00:23:02,000 --> 00:23:03,830 and its negative connotations. 569 00:23:06,290 --> 00:23:08,359 So the last term I'm 570 00:23:08,360 --> 00:23:10,429 going to have a look at is park 571 00:23:10,430 --> 00:23:13,039 data, and 572 00:23:13,040 --> 00:23:15,049 once again, I consult the dictionary to 573 00:23:15,050 --> 00:23:17,809 find out more about this word. 574 00:23:17,810 --> 00:23:21,049 So the Oxford English Dictionary 575 00:23:21,050 --> 00:23:23,509 is defined as a great a considerable 576 00:23:23,510 --> 00:23:25,399 volume mass. 577 00:23:25,400 --> 00:23:26,930 The collective mass of an object 578 00:23:28,040 --> 00:23:30,109 is usually used in the 579 00:23:30,110 --> 00:23:32,419 context of cargo and commerce 580 00:23:32,420 --> 00:23:34,549 in bog lying loose 581 00:23:34,550 --> 00:23:36,769 and heaps without a package 582 00:23:36,770 --> 00:23:38,179 and so on. 583 00:23:38,180 --> 00:23:40,069 So let's have a look at the concrete 584 00:23:40,070 --> 00:23:41,070 answers. 585 00:23:41,660 --> 00:23:43,429 The ability to acquire Bogda is 586 00:23:43,430 --> 00:23:44,569 necessary. 587 00:23:44,570 --> 00:23:46,909 The ability to collect data 588 00:23:46,910 --> 00:23:48,229 is essential. 589 00:23:48,230 --> 00:23:50,329 The bill supervises entirely 590 00:23:50,330 --> 00:23:53,179 the ability to collect bulk data, 591 00:23:53,180 --> 00:23:55,009 despite his attempts to conceal his 592 00:23:55,010 --> 00:23:56,299 activities. 593 00:23:56,300 --> 00:23:58,399 The agencies were able to use the data to 594 00:23:58,400 --> 00:24:00,469 identify that he had recently traveled 595 00:24:00,470 --> 00:24:01,700 to a European country. 596 00:24:02,840 --> 00:24:04,939 It empowers our services to access 597 00:24:04,940 --> 00:24:07,309 and analyze data, a tool 598 00:24:07,310 --> 00:24:09,259 that has become more important than ever 599 00:24:09,260 --> 00:24:10,459 before. 600 00:24:10,460 --> 00:24:12,769 So the context of bulk 601 00:24:12,770 --> 00:24:14,989 data can be split up into 602 00:24:14,990 --> 00:24:16,739 three major sections. 603 00:24:16,740 --> 00:24:19,069 The majority of the conferences, 604 00:24:19,070 --> 00:24:21,949 as we've seen in this in this excerpt, 605 00:24:21,950 --> 00:24:23,659 stresses the necessity and 606 00:24:23,660 --> 00:24:26,139 proportionality of data. 607 00:24:26,140 --> 00:24:28,399 A common strategy is to list 608 00:24:28,400 --> 00:24:30,649 incidents of when and how 609 00:24:30,650 --> 00:24:32,080 big data proved useful, 610 00:24:33,680 --> 00:24:35,839 16 percent referred to by 611 00:24:35,840 --> 00:24:38,569 date on a rather neutral contract text. 612 00:24:38,570 --> 00:24:41,329 That means stating amendments 613 00:24:41,330 --> 00:24:43,759 which refer to bulk data. 614 00:24:43,760 --> 00:24:46,099 And only 12 percent 615 00:24:46,100 --> 00:24:48,289 mentioned Baghdad on the critical 616 00:24:48,290 --> 00:24:51,019 context and exemplify 617 00:24:51,020 --> 00:24:53,329 this really abstract term 618 00:24:53,330 --> 00:24:55,309 and its indiscriminate approach by 619 00:24:55,310 --> 00:24:57,739 referring to 620 00:24:57,740 --> 00:24:59,369 a massive combine harvester. 621 00:25:01,100 --> 00:25:03,409 But of course, such 622 00:25:03,410 --> 00:25:05,929 using such vivid imagery is 623 00:25:05,930 --> 00:25:06,930 undesired. 624 00:25:07,820 --> 00:25:08,820 So 625 00:25:10,130 --> 00:25:11,689 let's have a look at the following quote. 626 00:25:12,800 --> 00:25:14,539 I appreciate that both powers are 627 00:25:14,540 --> 00:25:17,179 controversial, but I am absolutely 628 00:25:17,180 --> 00:25:20,089 sure that we do not conduct mass, that 629 00:25:20,090 --> 00:25:22,219 we do not conduct data harvesting in 630 00:25:22,220 --> 00:25:23,119 this country. 631 00:25:23,120 --> 00:25:24,889 It simply does not happen. 632 00:25:24,890 --> 00:25:26,989 The use of powers is not for 633 00:25:26,990 --> 00:25:29,119 that purpose, but for the purpose 634 00:25:29,120 --> 00:25:31,339 of examining material, even 635 00:25:31,340 --> 00:25:34,249 though that may be done and all 636 00:25:34,250 --> 00:25:36,259 it is done in a way that does not amount 637 00:25:36,260 --> 00:25:38,389 to the generalized harvesting of data 638 00:25:38,390 --> 00:25:39,979 for the extermination. 639 00:25:39,980 --> 00:25:42,139 It simply is not, 640 00:25:42,140 --> 00:25:43,279 says Mr Greif. 641 00:25:45,440 --> 00:25:47,719 So the Speaker does 642 00:25:47,720 --> 00:25:50,449 not clearly define what distinguishes 643 00:25:50,450 --> 00:25:52,519 data harvesting from the 644 00:25:52,520 --> 00:25:54,679 use of powers 645 00:25:54,680 --> 00:25:57,559 for the purpose of examining material 646 00:25:57,560 --> 00:25:58,560 in bulk. 647 00:26:00,320 --> 00:26:01,320 So 648 00:26:02,420 --> 00:26:03,709 everybody knows this man. 649 00:26:06,800 --> 00:26:08,899 And he talks about 650 00:26:08,900 --> 00:26:11,349 double thing in the following way, 651 00:26:11,350 --> 00:26:13,489 the power of holding two contradictory 652 00:26:13,490 --> 00:26:16,309 beliefs in one's mind simultaneously 653 00:26:16,310 --> 00:26:18,859 and accepting both of them. 654 00:26:18,860 --> 00:26:21,019 So what does this tell 655 00:26:21,020 --> 00:26:22,099 us? 656 00:26:22,100 --> 00:26:24,169 This unshakable faith in 657 00:26:24,170 --> 00:26:26,659 the good intentions of mass surveillance 658 00:26:26,660 --> 00:26:28,879 makes people actually believe that 659 00:26:28,880 --> 00:26:30,889 measures which correspond to the 660 00:26:30,890 --> 00:26:32,989 definition of mass surveillance 661 00:26:32,990 --> 00:26:34,579 are actually not mass surveillance. 662 00:26:36,980 --> 00:26:39,349 So maybe there 663 00:26:39,350 --> 00:26:41,449 are other things to worry about 664 00:26:41,450 --> 00:26:43,550 than Russian. But I don't know. 665 00:26:45,770 --> 00:26:47,899 Um, there's a general tendency 666 00:26:47,900 --> 00:26:50,479 of redefining mass surveillance 667 00:26:50,480 --> 00:26:52,519 as part data collection. 668 00:26:52,520 --> 00:26:54,859 And this aims at redirecting 669 00:26:54,860 --> 00:26:57,829 attention from mass to data, 670 00:26:57,830 --> 00:27:00,199 which means, in other words, 671 00:27:00,200 --> 00:27:02,419 it directs attention from a tomb 672 00:27:02,420 --> 00:27:04,639 which denotes a large group of people or 673 00:27:04,640 --> 00:27:06,739 items to a term which 674 00:27:06,740 --> 00:27:08,939 denotes something very abstract and 675 00:27:08,940 --> 00:27:09,989 depersonalized. 676 00:27:11,060 --> 00:27:12,060 So. 677 00:27:12,870 --> 00:27:14,939 Replacing a surveillance with 678 00:27:14,940 --> 00:27:17,069 collection further disguises 679 00:27:17,070 --> 00:27:19,409 the ongoing process collectors 680 00:27:19,410 --> 00:27:21,369 are very neutral and fuzzy term. 681 00:27:21,370 --> 00:27:23,729 Just think about people collecting stamps 682 00:27:23,730 --> 00:27:25,829 or coins or stuff like 683 00:27:25,830 --> 00:27:28,349 this where surveillance 684 00:27:28,350 --> 00:27:30,449 is very straightforward 685 00:27:30,450 --> 00:27:31,620 and unambiguous. 686 00:27:32,790 --> 00:27:34,529 Bug or in bulk. 687 00:27:34,530 --> 00:27:36,719 Maybe you remember the definition is 688 00:27:36,720 --> 00:27:39,599 usually used to denote merchandise, 689 00:27:39,600 --> 00:27:41,999 and this thinking further emphasizes 690 00:27:42,000 --> 00:27:43,200 depersonalization. 691 00:27:46,700 --> 00:27:48,979 It is constantly stressed in the 692 00:27:48,980 --> 00:27:51,049 Investigatory Powers Act 693 00:27:51,050 --> 00:27:53,599 for mass surveillance and 694 00:27:53,600 --> 00:27:56,659 data collection are not to be equated. 695 00:27:56,660 --> 00:27:58,879 Collection of data, most 696 00:27:58,880 --> 00:28:01,129 of which I never even read, 697 00:28:01,130 --> 00:28:03,579 does not constitute mass surveillance. 698 00:28:03,580 --> 00:28:04,580 Oh. 699 00:28:06,300 --> 00:28:09,689 Unfortunately, we cannot 700 00:28:09,690 --> 00:28:12,389 grasp from the debates what constitutes 701 00:28:12,390 --> 00:28:14,849 mass surveillance, if not the collection 702 00:28:14,850 --> 00:28:16,319 of communications data. 703 00:28:16,320 --> 00:28:17,320 And Doug. 704 00:28:19,900 --> 00:28:22,839 And this also 705 00:28:22,840 --> 00:28:25,179 throughout the entire corpus, there's 706 00:28:25,180 --> 00:28:27,519 a prevalence to refer 707 00:28:27,520 --> 00:28:29,799 to the investigatory powers bill 708 00:28:29,800 --> 00:28:32,049 and its measures in terms of what 709 00:28:32,050 --> 00:28:34,119 they are then, 710 00:28:34,120 --> 00:28:36,069 rather than saying what they actually 711 00:28:36,070 --> 00:28:36,909 are. 712 00:28:36,910 --> 00:28:38,979 So the strongest color 713 00:28:38,980 --> 00:28:41,739 codes of data are harvesting 714 00:28:41,740 --> 00:28:43,659 and and collection. 715 00:28:43,660 --> 00:28:45,430 So we now 716 00:28:46,900 --> 00:28:49,239 have a look at the concrete 717 00:28:49,240 --> 00:28:51,309 answers in news on the 718 00:28:51,310 --> 00:28:52,310 Web. 719 00:28:53,650 --> 00:28:55,749 A year ago, a presidential 720 00:28:55,750 --> 00:28:58,059 advisory board and privacy concluded 721 00:28:58,060 --> 00:29:00,459 that the bulk data collection was illegal 722 00:29:00,460 --> 00:29:01,480 and unproductive 723 00:29:03,070 --> 00:29:05,619 and business already faced 724 00:29:05,620 --> 00:29:08,019 by data collection of a different 725 00:29:08,020 --> 00:29:10,359 kind, as demonstrated by the hack 726 00:29:10,360 --> 00:29:12,099 and theft of millions of talk to our 727 00:29:12,100 --> 00:29:13,479 customers. 728 00:29:13,480 --> 00:29:14,859 If you're looking for a clear and 729 00:29:14,860 --> 00:29:17,859 comprehensive guide to how communications 730 00:29:17,860 --> 00:29:19,959 have been intercepted from 731 00:29:19,960 --> 00:29:22,179 cable cutting in the First World War to 732 00:29:22,180 --> 00:29:24,279 bulk data collection exposed 733 00:29:24,280 --> 00:29:25,960 by Ed Snowden, this is it. 734 00:29:27,040 --> 00:29:29,379 I have used the recent attacks to argue 735 00:29:29,380 --> 00:29:31,569 for broader surveillance and bulk 736 00:29:31,570 --> 00:29:33,789 data gathering bulk data 737 00:29:33,790 --> 00:29:35,919 collection, neither lawful nor 738 00:29:35,920 --> 00:29:36,879 effective. 739 00:29:36,880 --> 00:29:38,979 So as you can see in the context of 740 00:29:38,980 --> 00:29:41,419 bulk data and the news on the Web, corpus 741 00:29:41,420 --> 00:29:43,659 is very diverse. 742 00:29:43,660 --> 00:29:45,999 The majority of concurrences refers 743 00:29:46,000 --> 00:29:48,489 to the Snowden disclosures, the USA 744 00:29:48,490 --> 00:29:50,079 Freedom Act. 745 00:29:50,080 --> 00:29:52,239 Some articles express a very critical 746 00:29:52,240 --> 00:29:54,669 attitude and actually 747 00:29:54,670 --> 00:29:56,739 equate bulk data with mass 748 00:29:56,740 --> 00:29:58,900 surveillance is a quote 749 00:30:00,010 --> 00:30:02,259 about data retention, what anyone 750 00:30:02,260 --> 00:30:03,819 else would call mass surveillance. 751 00:30:05,770 --> 00:30:08,259 But the collection is referred to as 752 00:30:08,260 --> 00:30:10,119 the lawful more effective. 753 00:30:10,120 --> 00:30:11,550 Also seen that from the excerpts, 754 00:30:13,300 --> 00:30:15,609 several articles highlight the utility of 755 00:30:15,610 --> 00:30:16,749 data. 756 00:30:16,750 --> 00:30:17,750 Here's a quote. 757 00:30:18,400 --> 00:30:20,799 GQ analyst trawling bulk 758 00:30:20,800 --> 00:30:22,959 data spotted suspicious activity 759 00:30:22,960 --> 00:30:24,790 which turned out to be a Syrian jihadi. 760 00:30:26,530 --> 00:30:28,839 So only a very small number. 761 00:30:28,840 --> 00:30:31,479 Two out of 100 concurrences 762 00:30:31,480 --> 00:30:33,879 of the articles refer directly 763 00:30:33,880 --> 00:30:35,919 to the Investigatory Powers Bill. 764 00:30:35,920 --> 00:30:37,450 And this actually surprised me. 765 00:30:39,340 --> 00:30:41,529 The strongest color codes were to 766 00:30:41,530 --> 00:30:43,149 HQ and NSA. 767 00:30:43,150 --> 00:30:45,219 And once again, 768 00:30:45,220 --> 00:30:47,319 this indicates that bulk data 769 00:30:47,320 --> 00:30:49,839 is linked to modern governments. 770 00:30:49,840 --> 00:30:52,789 Similarly, as mass surveillance, 771 00:30:52,790 --> 00:30:55,029 so disregarding 772 00:30:55,030 --> 00:30:56,030 acronyms 773 00:30:57,880 --> 00:31:00,489 outside most strongly colocated 774 00:31:00,490 --> 00:31:02,679 with interception, collection and 775 00:31:02,680 --> 00:31:03,680 interference. 776 00:31:08,390 --> 00:31:10,489 OK, in 777 00:31:10,490 --> 00:31:12,379 the Investigatory Powers Act, the 778 00:31:12,380 --> 00:31:15,169 majority of concrete answers highlighted 779 00:31:15,170 --> 00:31:17,209 the importance and necessity of bulk 780 00:31:17,210 --> 00:31:18,379 data. 781 00:31:18,380 --> 00:31:20,659 There is in the newspaper Corpuz, 782 00:31:22,190 --> 00:31:24,439 the majority of concurrences 783 00:31:24,440 --> 00:31:26,929 referred to the Snowden revelations 784 00:31:26,930 --> 00:31:28,340 are the USA Freedom Act. 785 00:31:29,690 --> 00:31:32,419 And in the parliamentary debates, 786 00:31:32,420 --> 00:31:34,579 we have this tendency of redefining 787 00:31:34,580 --> 00:31:36,979 mass surveillance as part data 788 00:31:36,980 --> 00:31:38,919 or bulk collection. 789 00:31:38,920 --> 00:31:41,149 OK, so 790 00:31:41,150 --> 00:31:42,499 let's give a short summary. 791 00:31:45,500 --> 00:31:47,869 Snooper's charter and mass surveillance, 792 00:31:47,870 --> 00:31:50,179 where the two keywords 793 00:31:50,180 --> 00:31:52,339 which were used most 794 00:31:52,340 --> 00:31:54,409 diversely and they were used in 795 00:31:54,410 --> 00:31:56,209 two completely different ways. 796 00:31:56,210 --> 00:31:57,229 And the two corpora, 797 00:31:58,370 --> 00:32:00,229 super shadow and mass surveillance were 798 00:32:00,230 --> 00:32:02,299 also the two least 799 00:32:02,300 --> 00:32:04,849 frequent keywords in the Investigatory 800 00:32:04,850 --> 00:32:06,919 Powers Bill and the 801 00:32:06,920 --> 00:32:09,289 two most frequent keywords 802 00:32:09,290 --> 00:32:10,919 in the news on the Web corpus. 803 00:32:12,770 --> 00:32:15,079 So as I pointed 804 00:32:15,080 --> 00:32:17,689 out before, there's a general prevalence 805 00:32:17,690 --> 00:32:19,849 to refer to the Investigatory 806 00:32:19,850 --> 00:32:22,039 Powers Act and as measures in 807 00:32:22,040 --> 00:32:23,839 terms of what they're not. 808 00:32:23,840 --> 00:32:25,909 And as a consequence, the 809 00:32:25,910 --> 00:32:28,009 debates are no longer concerned 810 00:32:28,010 --> 00:32:30,169 with the nature of these 811 00:32:30,170 --> 00:32:32,539 things because things that only defined 812 00:32:32,540 --> 00:32:33,540 X negatively 813 00:32:35,420 --> 00:32:37,609 so and 814 00:32:37,610 --> 00:32:40,099 the Investigatory Powers Act, you have 815 00:32:40,100 --> 00:32:42,419 a strong tendency for rather non 816 00:32:42,420 --> 00:32:43,490 descriptive terms. 817 00:32:44,660 --> 00:32:47,629 But in the news, in the Web corpus, 818 00:32:47,630 --> 00:32:49,939 there is a stronger prevalence 819 00:32:49,940 --> 00:32:52,099 for negatively connoted words such as 820 00:32:52,100 --> 00:32:54,619 mass surveillance, and this might 821 00:32:54,620 --> 00:32:57,529 be rooted in the nature of the two 822 00:32:57,530 --> 00:32:59,359 spheres. I had a look at 823 00:33:00,770 --> 00:33:03,319 a newspaper has an interest 824 00:33:03,320 --> 00:33:05,629 in translating legal language 825 00:33:05,630 --> 00:33:08,779 into language its readers understand, 826 00:33:08,780 --> 00:33:10,699 whereas political language is often 827 00:33:10,700 --> 00:33:12,829 designed to disguise 828 00:33:12,830 --> 00:33:13,830 unpleasant facts. 829 00:33:15,360 --> 00:33:17,689 So that's 830 00:33:17,690 --> 00:33:18,690 Paternalists. 831 00:33:19,910 --> 00:33:21,739 I did a lot of reading while I was 832 00:33:21,740 --> 00:33:23,899 compiling this analysis and I love 833 00:33:23,900 --> 00:33:26,059 reading through all sorts of 834 00:33:26,060 --> 00:33:28,129 debates. And I compiled a top 835 00:33:28,130 --> 00:33:30,619 three of my favorite quotes 836 00:33:30,620 --> 00:33:32,719 from these debates when 837 00:33:32,720 --> 00:33:33,720 you already know. 838 00:33:34,880 --> 00:33:36,949 So I sincerely hope that as the book 839 00:33:36,950 --> 00:33:39,079 proceeds, we have a way to go. 840 00:33:39,080 --> 00:33:41,239 Yet we will explain that we 841 00:33:41,240 --> 00:33:43,279 do not conduct mass surveillance in the 842 00:33:43,280 --> 00:33:44,359 UK. 843 00:33:44,360 --> 00:33:46,809 Indeed, it is not done in the USA. 844 00:33:48,800 --> 00:33:50,899 Collection of data, most 845 00:33:50,900 --> 00:33:53,179 of which I never even read, does 846 00:33:53,180 --> 00:33:54,859 not constitute mass surveillance. 847 00:34:01,390 --> 00:34:04,129 Oh, yeah. 848 00:34:04,130 --> 00:34:05,130 And second one, 849 00:34:06,200 --> 00:34:08,329 however uneasy we may feel 850 00:34:08,330 --> 00:34:10,408 about Internet connection records 851 00:34:10,409 --> 00:34:12,589 or thematic warrants, 852 00:34:12,590 --> 00:34:14,869 that does not compare to the infinitely 853 00:34:14,870 --> 00:34:16,939 greater unease we also feel 854 00:34:16,940 --> 00:34:19,669 about our intelligence agencies 855 00:34:19,670 --> 00:34:21,829 being unable to use those tools to 856 00:34:21,830 --> 00:34:22,830 keep us safe. 857 00:34:26,670 --> 00:34:28,359 And once again, most of 858 00:34:29,550 --> 00:34:31,859 the power of holding two contradictory 859 00:34:31,860 --> 00:34:34,709 beliefs in one's mind simultaneously 860 00:34:34,710 --> 00:34:36,809 and accepting both of them 861 00:34:36,810 --> 00:34:39,209 to tell deliberate lies while genuinely 862 00:34:39,210 --> 00:34:41,279 believing in them to forget 863 00:34:41,280 --> 00:34:43,829 any fact that has become inconvenient. 864 00:34:43,830 --> 00:34:45,988 And then when it becomes necessary 865 00:34:45,989 --> 00:34:48,509 again to throw it back from oblivion 866 00:34:48,510 --> 00:34:50,729 for just as long as it is needed 867 00:34:50,730 --> 00:34:53,099 to deny the existence of objective 868 00:34:53,100 --> 00:34:55,169 reality and all the while to 869 00:34:55,170 --> 00:34:57,269 take account of the reality which 870 00:34:57,270 --> 00:34:59,939 one denies all this is indispensably 871 00:34:59,940 --> 00:35:00,940 necessary 872 00:35:02,100 --> 00:35:03,100 and. 873 00:35:04,930 --> 00:35:07,869 Appreciate the bypass, a controversial, 874 00:35:07,870 --> 00:35:09,789 but I'm absolutely sure that we do not 875 00:35:09,790 --> 00:35:12,579 conduct data harvesting in this country, 876 00:35:12,580 --> 00:35:13,769 it simply does not happen. 877 00:35:14,830 --> 00:35:16,809 The use of powers is not for that 878 00:35:16,810 --> 00:35:19,059 purpose, but for the purpose of 879 00:35:19,060 --> 00:35:21,309 examining material, even 880 00:35:21,310 --> 00:35:24,069 though that may be done in bulk. 881 00:35:24,070 --> 00:35:26,379 It is done in a way that does not amount 882 00:35:26,380 --> 00:35:28,509 to the generalized harvesting of data 883 00:35:28,510 --> 00:35:30,219 for the extermination. 884 00:35:30,220 --> 00:35:31,220 It simply is not. 885 00:35:32,870 --> 00:35:35,119 OK, so 886 00:35:35,120 --> 00:35:37,309 a few words on the bibliography, 887 00:35:37,310 --> 00:35:39,739 if you're really interested in Curbers 888 00:35:39,740 --> 00:35:41,959 linguistics, I would recommend 889 00:35:41,960 --> 00:35:44,449 Mr. Paul Baker and 890 00:35:44,450 --> 00:35:46,609 pretty much everything he has written. 891 00:35:46,610 --> 00:35:48,649 And if you want to play around with the 892 00:35:48,650 --> 00:35:50,809 corpus is 893 00:35:50,810 --> 00:35:52,999 you can find it online now. 894 00:35:53,000 --> 00:35:54,000 Corpus. 895 00:35:55,950 --> 00:35:57,989 OK, any questions? 896 00:36:07,840 --> 00:36:09,899 Yes, please come to the 897 00:36:09,900 --> 00:36:10,900 microphones 898 00:36:12,470 --> 00:36:14,839 and I will tell you this first, and 899 00:36:14,840 --> 00:36:17,119 so we start with microphone number one. 900 00:36:17,120 --> 00:36:18,619 Thank you for the interesting. 901 00:36:18,620 --> 00:36:20,959 What did you observe 902 00:36:20,960 --> 00:36:23,479 a change in time 903 00:36:23,480 --> 00:36:26,029 of the usage of words. 904 00:36:26,030 --> 00:36:28,189 So this is basically is Tukaram 905 00:36:28,190 --> 00:36:30,529 of the of the usage, but 906 00:36:30,530 --> 00:36:32,629 something like the Times to in 907 00:36:32,630 --> 00:36:35,539 this. So we had in Germany 908 00:36:35,540 --> 00:36:37,699 the voting for start 909 00:36:37,700 --> 00:36:40,039 to save something for later, which 910 00:36:40,040 --> 00:36:42,439 is has a positive 911 00:36:42,440 --> 00:36:43,489 connotation. 912 00:36:43,490 --> 00:36:46,219 But in in in the discussion 913 00:36:46,220 --> 00:36:48,379 it came that everybody knew 914 00:36:48,380 --> 00:36:50,899 for a fact which is bad, 915 00:36:50,900 --> 00:36:53,329 and then they changed 916 00:36:53,330 --> 00:36:54,330 the voting. 917 00:36:56,240 --> 00:36:58,789 So you mean in this discussion? 918 00:36:58,790 --> 00:37:00,859 From what I know, the terminology has 919 00:37:00,860 --> 00:37:02,119 not changed. 920 00:37:02,120 --> 00:37:04,549 So I did not take a closer 921 00:37:04,550 --> 00:37:07,279 look at this first snooper's charter 922 00:37:07,280 --> 00:37:09,619 because the second snapshot 923 00:37:09,620 --> 00:37:11,119 of the one I was talking about is a 924 00:37:11,120 --> 00:37:13,439 milder version of the first supercharger. 925 00:37:13,440 --> 00:37:15,619 But from what I know, the terminology has 926 00:37:15,620 --> 00:37:17,749 not changed. It was always called 927 00:37:17,750 --> 00:37:20,179 bulk data, but data collection 928 00:37:20,180 --> 00:37:23,449 by data interception, stuff like this. 929 00:37:23,450 --> 00:37:24,709 Thank you. 930 00:37:24,710 --> 00:37:27,289 OK, question number three, 931 00:37:27,290 --> 00:37:29,449 when you searched 932 00:37:29,450 --> 00:37:32,119 in the law 933 00:37:32,120 --> 00:37:34,129 in the last, you search for the term 934 00:37:34,130 --> 00:37:36,979 Investigatory Powers Act. 935 00:37:36,980 --> 00:37:38,389 But when you search in the online 936 00:37:38,390 --> 00:37:40,009 newspaper, you search for snooper's 937 00:37:40,010 --> 00:37:42,409 charter. Might that have introduced 938 00:37:42,410 --> 00:37:44,329 some kind of bias? 939 00:37:44,330 --> 00:37:45,980 The thing is, if you 940 00:37:47,090 --> 00:37:48,320 if you can solve this 941 00:37:49,550 --> 00:37:51,679 hanser, this documentation of 942 00:37:51,680 --> 00:37:52,680 parliamentary debates, 943 00:37:53,930 --> 00:37:56,149 the debates are only 944 00:37:56,150 --> 00:37:58,519 found if you enter the 945 00:37:58,520 --> 00:37:59,520 official term. 946 00:38:00,440 --> 00:38:02,749 Yeah, but the thing is 947 00:38:02,750 --> 00:38:03,800 that, you 948 00:38:04,860 --> 00:38:07,579 know, if you 949 00:38:07,580 --> 00:38:09,109 look for newspaper articles. 950 00:38:15,470 --> 00:38:17,809 Sorry, can you ask the question again? 951 00:38:17,810 --> 00:38:20,029 So basically, have you considered 952 00:38:20,030 --> 00:38:22,279 Searching for Investigatory Powers Act 953 00:38:22,280 --> 00:38:24,199 and the newspapers as well to see whether 954 00:38:25,790 --> 00:38:27,469 these articles maybe have a different 955 00:38:27,470 --> 00:38:29,749 language about it than the 956 00:38:29,750 --> 00:38:31,639 articles containing Snooper's Charter, 957 00:38:31,640 --> 00:38:33,769 which is a loaded term, I guess? 958 00:38:33,770 --> 00:38:35,850 Yeah, uh, no, I didn't do that. 959 00:38:38,330 --> 00:38:40,339 OK, no one. 960 00:38:40,340 --> 00:38:42,649 Yeah. Say I campaigned against 961 00:38:42,650 --> 00:38:44,829 the powers that be as part 962 00:38:44,830 --> 00:38:46,669 as a member of Overwrites Rights Group. 963 00:38:46,670 --> 00:38:48,739 And I was just wondering how 964 00:38:48,740 --> 00:38:50,569 when we had the press very much on our 965 00:38:50,570 --> 00:38:52,789 side, how you recommends 966 00:38:52,790 --> 00:38:55,249 that we can actually improve 967 00:38:55,250 --> 00:38:56,449 like we lost that one. 968 00:38:59,020 --> 00:39:00,489 That quick, yes, 969 00:39:01,630 --> 00:39:03,739 yes, so I campaigned with 970 00:39:03,740 --> 00:39:06,009 open wide sweep against the steepest 971 00:39:06,010 --> 00:39:08,499 charter, and we mostly had the press 972 00:39:08,500 --> 00:39:10,599 on our side, but we still lost. 973 00:39:12,020 --> 00:39:13,509 I just wondered if you have any advice 974 00:39:13,510 --> 00:39:14,949 for how we could do better. 975 00:39:17,830 --> 00:39:19,179 It's a very tricky question, 976 00:39:21,790 --> 00:39:24,129 but I found many articles saying 977 00:39:24,130 --> 00:39:26,229 that many 978 00:39:26,230 --> 00:39:28,479 UK citizens don't 979 00:39:28,480 --> 00:39:30,819 even know what the Snooper's Charter 980 00:39:30,820 --> 00:39:33,579 is. And I think 981 00:39:33,580 --> 00:39:35,199 a lot was done to. 982 00:39:35,200 --> 00:39:37,329 I mean, these debates have been have 983 00:39:37,330 --> 00:39:39,319 been going on for almost two years. 984 00:39:39,320 --> 00:39:41,139 I think a lot has been done to 985 00:39:42,520 --> 00:39:44,709 disguise the ongoing process of what 986 00:39:44,710 --> 00:39:46,300 is actually being discussed. 987 00:39:48,600 --> 00:39:49,600 But I 988 00:39:50,710 --> 00:39:53,049 you're analyzing something afterwards, 989 00:39:53,050 --> 00:39:54,759 the same a suggestion how to do it 990 00:39:54,760 --> 00:39:56,919 better. But, yeah, it's 991 00:39:56,920 --> 00:39:59,079 a nice idea to to perhaps use 992 00:39:59,080 --> 00:40:01,209 this as a basis for next time to make 993 00:40:01,210 --> 00:40:02,210 analysis this way. 994 00:40:03,340 --> 00:40:05,109 So we come to you. 995 00:40:05,110 --> 00:40:07,419 We have also a question from the 996 00:40:07,420 --> 00:40:09,729 signal from the Internet. 997 00:40:09,730 --> 00:40:12,279 Yes, we do what we're 998 00:40:12,280 --> 00:40:14,799 told, but did you use 999 00:40:14,800 --> 00:40:17,079 for the collection and analysis 1000 00:40:17,080 --> 00:40:18,610 and now they are available 1001 00:40:19,990 --> 00:40:21,339 for the corpus? 1002 00:40:21,340 --> 00:40:22,509 I can compiled myself. 1003 00:40:22,510 --> 00:40:24,579 I use open source 1004 00:40:24,580 --> 00:40:26,769 tool called and conc 1005 00:40:26,770 --> 00:40:27,999 a. 1006 00:40:28,000 --> 00:40:29,859 CEO and see. 1007 00:40:29,860 --> 00:40:32,109 And if you search for it, you can find 1008 00:40:32,110 --> 00:40:33,879 it and download it. 1009 00:40:33,880 --> 00:40:36,519 And the web based corpus 1010 00:40:36,520 --> 00:40:38,889 comes with a built in tool. 1011 00:40:38,890 --> 00:40:41,019 You just enter the search term, 1012 00:40:41,020 --> 00:40:43,569 you're interested it on the website 1013 00:40:43,570 --> 00:40:46,119 and it has its own 1014 00:40:46,120 --> 00:40:48,219 analysis tool. 1015 00:40:48,220 --> 00:40:50,739 But there are all sorts of concordance 1016 00:40:50,740 --> 00:40:51,819 programs. 1017 00:40:51,820 --> 00:40:53,889 I think the most known one is called 1018 00:40:53,890 --> 00:40:56,139 wordsmith, but it's 1019 00:40:56,140 --> 00:40:58,409 not free from what I know. 1020 00:40:58,410 --> 00:40:59,919 OK, OK. 1021 00:40:59,920 --> 00:41:00,969 Microphone number one, 1022 00:41:02,200 --> 00:41:03,819 thank you for your brilliant analysis. 1023 00:41:03,820 --> 00:41:04,809 No, sir. 1024 00:41:04,810 --> 00:41:06,579 My first part of the question would have 1025 00:41:06,580 --> 00:41:07,959 been similar to the last one. 1026 00:41:07,960 --> 00:41:10,179 So what narratives can we use to 1027 00:41:10,180 --> 00:41:12,579 hack the discourse on such 1028 00:41:12,580 --> 00:41:14,559 surveillance questions? 1029 00:41:14,560 --> 00:41:16,899 But yeah, and the second 1030 00:41:16,900 --> 00:41:19,029 part of my question is maybe 1031 00:41:19,030 --> 00:41:21,519 a kind of another term 1032 00:41:21,520 --> 00:41:23,619 that was very interesting in the 1033 00:41:23,620 --> 00:41:25,719 discourse security, because 1034 00:41:25,720 --> 00:41:28,179 security most often is used in a way 1035 00:41:28,180 --> 00:41:30,339 that it is like there 1036 00:41:30,340 --> 00:41:31,359 is a terrorist attack. 1037 00:41:31,360 --> 00:41:32,319 So we need more security. 1038 00:41:32,320 --> 00:41:33,669 So we need more surveillance. 1039 00:41:33,670 --> 00:41:35,499 But it can also be used in a way of 1040 00:41:36,720 --> 00:41:38,889 the Domestic Battery Powers 1041 00:41:38,890 --> 00:41:41,109 Act introduces restrictions on 1042 00:41:41,110 --> 00:41:43,239 encryption, which we can say to 1043 00:41:43,240 --> 00:41:45,339 security in our all collective 1044 00:41:45,340 --> 00:41:46,929 security and the Internet. 1045 00:41:46,930 --> 00:41:49,059 So, yeah, 1046 00:41:49,060 --> 00:41:51,039 maybe I would be interested in the use of 1047 00:41:51,040 --> 00:41:52,449 security in this particular 1048 00:41:54,190 --> 00:41:55,149 particular discourse. 1049 00:41:55,150 --> 00:41:57,529 Thank you. OK, thank you. 1050 00:41:57,530 --> 00:41:59,679 Um, well, as 1051 00:41:59,680 --> 00:42:01,299 I said before, if you have such a large 1052 00:42:01,300 --> 00:42:04,239 corpus, you can run endless analysis. 1053 00:42:04,240 --> 00:42:07,029 Uh, I thought about doing security. 1054 00:42:07,030 --> 00:42:09,609 One thing that is always 1055 00:42:09,610 --> 00:42:11,619 say, um, 1056 00:42:13,060 --> 00:42:15,219 it was not as striking 1057 00:42:15,220 --> 00:42:17,319 because it is used over and 1058 00:42:17,320 --> 00:42:18,729 over again. 1059 00:42:18,730 --> 00:42:21,129 And I wanted to pick out 1060 00:42:21,130 --> 00:42:23,109 a more specific term. 1061 00:42:23,110 --> 00:42:25,299 But you could I mean, you could 1062 00:42:25,300 --> 00:42:27,069 focus on verbs. 1063 00:42:27,070 --> 00:42:28,960 They use adjectives. 1064 00:42:31,510 --> 00:42:33,639 I mean, if you if you're into 1065 00:42:33,640 --> 00:42:36,099 it, you can just search for it online 1066 00:42:36,100 --> 00:42:38,529 and do your own analysis 1067 00:42:38,530 --> 00:42:40,599 and just have a look. I mean, even if you 1068 00:42:40,600 --> 00:42:42,759 look at a couple of concrete answers, 1069 00:42:42,760 --> 00:42:44,919 you quite quickly get 1070 00:42:44,920 --> 00:42:47,049 an idea of how this word 1071 00:42:47,050 --> 00:42:48,050 is used and. 1072 00:42:50,580 --> 00:42:52,649 OK, so that's it 1073 00:42:52,650 --> 00:42:55,049 for tonight or for this talk? 1074 00:42:55,050 --> 00:42:57,029 I think there's another one afterwards. 1075 00:42:57,030 --> 00:42:59,279 So please let 1076 00:42:59,280 --> 00:43:01,379 us thank Lisa again for this amazing 1077 00:43:01,380 --> 00:43:02,660 talk and so nice to.