1 00:00:00,060 --> 00:00:01,260 Thank you all for coming. 2 00:00:01,260 --> 00:00:12,830 As someone who seldom turns out to lectures by others, I can greatly appreciate the trouble and the meaning of your turning up here, so thank you. 3 00:00:12,830 --> 00:00:22,670 It would be presumptuous of me in the extreme to connect the period since the inception of digital computing about 70 years ago, 4 00:00:22,670 --> 00:00:26,330 about as long as I've been around with the Enlightenment. 5 00:00:26,330 --> 00:00:38,120 But for the fact that apart from the noise and clutter that computing is has caused in spite of the deeply sinister uses to which it has been put, 6 00:00:38,120 --> 00:00:42,830 the central question it raises is the question of reason. 7 00:00:42,830 --> 00:00:51,800 What do we need to consider as scholars when we use computing to reason about our objects of study? 8 00:00:51,800 --> 00:00:59,660 How are its implementations of computing that we inherit and make ourselves changing what we do? 9 00:00:59,660 --> 00:01:04,250 Those are big questions I will sniff around the edges of. 10 00:01:04,250 --> 00:01:11,150 Now that leads me to to the Japanese print here to let you know what I will be. 11 00:01:11,150 --> 00:01:18,590 What you can expect from me. This tale the the hand of your son. 12 00:01:18,590 --> 00:01:25,610 Forgive the terrible Japanese pronunciation. In the version told by the roboticist Masahiro Mori, 13 00:01:25,610 --> 00:01:34,010 an old couple acquire a dog who runs into their garden, sniffing the ground stops and barks insistently. 14 00:01:34,010 --> 00:01:40,460 Dig here, dig here. The old man digs and finds a hoard of gold coins. 15 00:01:40,460 --> 00:01:44,450 Mori very self-deprecating. Li says. I'm that dog. 16 00:01:44,450 --> 00:01:49,820 I don't do any digging. I just run around and find interesting things and bark a lot. 17 00:01:49,820 --> 00:01:51,460 I'm not so self-deprecating. 18 00:01:51,460 --> 00:01:59,750 I mean it when I see that that's what I do is run through the field and bark what I find interesting stuff and that's what I'm going to do tonight. 19 00:01:59,750 --> 00:02:10,420 So profundity is perhaps not, but I hope you will be inspired by some of this to do some digging yourself. 20 00:02:10,420 --> 00:02:17,350 You will I am guessing you have heard of the literary historian Franco Moretti's term distant reading, 21 00:02:17,350 --> 00:02:26,140 by which he means the use of computational methods to find patterns in large bodies of text large textual corpora. 22 00:02:26,140 --> 00:02:29,290 In the epigraph to his first discussion of it, 23 00:02:29,290 --> 00:02:38,520 Moretti quotes Aaron's words from the libretto to Arnold Schoenberg's unfinished and quite unorthodox opera Moses and Aaron. 24 00:02:38,520 --> 00:02:45,420 This is what Aaron says. My mission to see it more simply than I understand it. 25 00:02:45,420 --> 00:02:50,520 So what Moretti is doing here is striking in an implicit analogy, on the one hand, 26 00:02:50,520 --> 00:03:00,000 to humanly comprehensible simplifications of the transcendent Aaron's Golden Calf and computational results, 27 00:03:00,000 --> 00:03:04,530 on the other hand, to rich complexities beyond comprehension. 28 00:03:04,530 --> 00:03:12,750 The idea of the divine and the whole of literature between them is the Negotiator Liben. 29 00:03:12,750 --> 00:03:21,570 It saw the relationship somewhat differently, not an external drama, but the ongoing internal negotiation within a happy marriage. 30 00:03:21,570 --> 00:03:28,650 Joining theoreticians and empiric this relationship between the dirt and the word. 31 00:03:28,650 --> 00:03:36,480 As Emily firmly said, once of archaeology and philology is where the promise of digital computing lies. 32 00:03:36,480 --> 00:03:42,870 My aim here is to question and explore that promise. 33 00:03:42,870 --> 00:03:48,510 To my mind, the negotiating in that metaphor is central. 34 00:03:48,510 --> 00:03:54,660 It shifts attention from the fact that computing and the humanities have intersected 35 00:03:54,660 --> 00:04:00,990 to the actions we take in that intersection loudness as metaphor of negotiation. 36 00:04:00,990 --> 00:04:07,020 The FĂ©lix Connubial is a provocative one for in a relationship of that kind. 37 00:04:07,020 --> 00:04:14,070 Each partner affects the other and is in turn affected in a manner of speaking the CoreValve. 38 00:04:14,070 --> 00:04:22,350 So also the developmental spiral in the history of technology, from invention to assimilation to new invention. 39 00:04:22,350 --> 00:04:33,870 The traffic between scholarship and computing is usually conceived and followed as if it were a one way street there that we must change. 40 00:04:33,870 --> 00:04:44,400 Perhaps it was labelmates or perhaps Voltaire, who was the last in the West to be capable of truly encyclopaedic knowledge in that sense, at least. 41 00:04:44,400 --> 00:04:48,480 This lecture sits well under both their names. Undoubtedly, 42 00:04:48,480 --> 00:04:58,860 the many specialisms into which the whole range of their concerns have branched and branched again and again are affected by the digital machine. 43 00:04:58,860 --> 00:05:08,520 But I want to argue for more than that for the encyclopaedic relevance of digital computing and scholarship to each other. 44 00:05:08,520 --> 00:05:13,200 And particularly because it is so undervalued by experts on both sides, 45 00:05:13,200 --> 00:05:19,800 computing's need for the rich inheritance and ongoing work of our literally human resources. 46 00:05:19,800 --> 00:05:27,570 I want someone to invite to entice as many bearers of green from as many academic silos as possible, 47 00:05:27,570 --> 00:05:30,690 including those silos that have been around for a long, 48 00:05:30,690 --> 00:05:41,160 long before digital computing was invented, so that our understanding of how to use it does not suffer from starvation of the mind and spirit. 49 00:05:41,160 --> 00:05:49,560 In a nutshell, this understanding shows us that the machine is not just an obedient servant to our whims, 50 00:05:49,560 --> 00:06:01,170 thoughts and questions by its resistance to us, as well as by its enabling amplification of our imaginative capacities. 51 00:06:01,170 --> 00:06:08,630 It has the potential of becoming a powerful companion with which to reason to question. 52 00:06:08,630 --> 00:06:20,860 And perhaps someday to converse. Allow me to enlarge the point if the computer were only a platform for useful applications. 53 00:06:20,860 --> 00:06:27,300 Acknowledged Jukebox, a filing system, a communications platform, digital typewriter and the like. 54 00:06:27,300 --> 00:06:29,730 Then there would be much less for me to talk about. 55 00:06:29,730 --> 00:06:37,920 In fact, I wouldn't be here as ubiquitous and central to our work as these applications are as much as they affect us. 56 00:06:37,920 --> 00:06:39,750 They aren't my subject, 57 00:06:39,750 --> 00:06:48,270 nor are the genuine accomplishments of digital scholarship brilliantly summarised by a Laura Mandel three weeks ago in a blog post. 58 00:06:48,270 --> 00:06:55,980 Experiencing the bust. Responding to a polemical rant in the American Chronicle of Higher Education, 59 00:06:55,980 --> 00:07:04,170 I won't repeat her insightful defence or add to her list of worthy studies, which are indeed growing in number and easy to find. 60 00:07:04,170 --> 00:07:09,480 As long as you know where to look and can separate the wheat from the chaff. 61 00:07:09,480 --> 00:07:14,220 Later on, we return to use of the machine for straightforward access to stuff, 62 00:07:14,220 --> 00:07:20,190 but throughout my lecture, my emphasis will be on the enlightening cognitive resistance. 63 00:07:20,190 --> 00:07:28,290 I just spoke of. And that kind of problematic liberation inherent to the machine from its beginnings. 64 00:07:28,290 --> 00:07:31,280 And I want to persuade you that the very, 65 00:07:31,280 --> 00:07:38,750 that very quickly such enquiry leads to the deep end in which the older disciplines have been from their beginnings, 66 00:07:38,750 --> 00:07:48,100 have been swimming and teaching others to swim. But since Professor Cronk invited me here, 67 00:07:48,100 --> 00:07:56,950 I must at least begin with a class of examples relevant to the Voltaire Foundation's ambitious editorial project. 68 00:07:56,950 --> 00:08:06,250 In his essay, Editing is a theoretical pursuit. Jerome again cites a number of remarkable scholarly editions emergent as he says, 69 00:08:06,250 --> 00:08:12,490 under a digital horizon, prophesying an electronic existence for themselves. 70 00:08:12,490 --> 00:08:14,470 That's the way Jerry writes. 71 00:08:14,470 --> 00:08:22,810 These courtesies, he continues, comprise our ages in cannabis books and winding sheets rather than swaddling clothes at once. 72 00:08:22,810 --> 00:08:27,670 Very beautiful and very ugly, fascinating and tedious. 73 00:08:27,670 --> 00:08:33,880 These books drive the resources of the codex to its limits and beyond. 74 00:08:33,880 --> 00:08:42,070 Hence, the urgent dream of the Digital Edition with its many tie encoded approximations. 75 00:08:42,070 --> 00:08:49,690 I don't certainly don't want to belittle or seem arrogantly to correct the industry of scholarship producing these editions. 76 00:08:49,690 --> 00:08:54,640 Rather, I want to talk about this industry's theoretical pursuit from a theoretical 77 00:08:54,640 --> 00:09:00,130 perspective relevant to all the disciplines of the arts and letters for which, 78 00:09:00,130 --> 00:09:05,470 by the way, I've co-opted the term literally human resources from the Oxford Classics programme. 79 00:09:05,470 --> 00:09:14,140 Again, my subject is the potential of the digital medium to respond to the emergent demands not only of the critical edition, 80 00:09:14,140 --> 00:09:23,090 but also all other scholarly terms forms of expression. 81 00:09:23,090 --> 00:09:30,770 The critic I.a Richards, who helped establish English literary criticism, various literary studies at Cambridge, 82 00:09:30,770 --> 00:09:37,640 began his book Principles of Literary Criticism, with the assertion that a book is a machine to think with. 83 00:09:37,640 --> 00:09:40,460 Richards compared his book to a loom. 84 00:09:40,460 --> 00:09:50,750 Perhaps not coincidentally, Jack Card's machine, whose control mechanisms Charles Babbage adopted for his analytical engine in 1836. 85 00:09:50,750 --> 00:09:59,810 With the digital edition in mind, I want to put you in mind of how the design of any such machine to think with from cuneiform 86 00:09:59,810 --> 00:10:06,710 tablets and papyrus rolls of Alexandria to the paperback shapes the thinkers cognitive paths. 87 00:10:06,710 --> 00:10:10,970 Consider these six examples. 88 00:10:10,970 --> 00:10:19,130 The first is a 19th century glossed manuscript of Marciano's Capellas Late Antique Work Day, new piece on the marriage of philology and Mercury, 89 00:10:19,130 --> 00:10:28,700 in which the glosses weave together traditional authorities and commentary with the fifth century text glosses here not only clarify, 90 00:10:28,700 --> 00:10:32,510 sometimes they obscure by encryption, wordplay, puzzles, 91 00:10:32,510 --> 00:10:44,710 allegories and etymology is paradoxically revealing by concealing the reader's path is often purposefully an intricate maze. 92 00:10:44,710 --> 00:10:55,360 The second one in the middle. Top middle is a verbal concordance, a 13, 12th or 13th 14th century manuscript, I think it says. 93 00:10:55,360 --> 00:11:00,550 But the concordance itself was invented in the late 12th or early 13th century. 94 00:11:00,550 --> 00:11:07,390 This research tool is there, and Krauss would say, directs the enquiring mind from a given word, from a given head word, 95 00:11:07,390 --> 00:11:13,630 which you can see illuminated in those columns to all of the passages in the wall 96 00:11:13,630 --> 00:11:22,630 gate where it is attested reading is to some degree randomised as well as directed. 97 00:11:22,630 --> 00:11:30,010 The third example is a 20th century but solidly traditional English Bible, published by Oxford University Press, 98 00:11:30,010 --> 00:11:35,500 whose central column links passages both within each testament and between them, 99 00:11:35,500 --> 00:11:39,610 not by the words they use, but according to the topological structure. 100 00:11:39,610 --> 00:11:47,320 They illuminate under the principle articulated by Saint Augustine for the Christian Bible in the old, he wrote. 101 00:11:47,320 --> 00:11:53,220 The new is concealed and in the new, the old is revealed. 102 00:11:53,220 --> 00:11:59,700 The fourth example of lower left is a computer generated keyword in context or quick concordance, 103 00:11:59,700 --> 00:12:07,350 whose format for stalls reading redirecting attention to the centred word to its nearest neighbours. 104 00:12:07,350 --> 00:12:17,190 Left and right. It is the basis for the field of of corpus linguistics, according to linguist Firth's dictum. 105 00:12:17,190 --> 00:12:23,250 You shall know a word by the company it keeps or is your mother would have said to you in your 106 00:12:23,250 --> 00:12:30,930 childhood if she had spoken Latin NASCAR to or a silky is the fifth example in the middle. 107 00:12:30,930 --> 00:12:43,160 A lower is Ramon Yarl's 14th century paper machine, designed to engage the reader combinatorial in correlating the terms of a spiritual logic. 108 00:12:43,160 --> 00:12:49,280 And the last is one of Herman Goldstein's and John Fund Newman's diagrams to indicate the logical 109 00:12:49,280 --> 00:12:55,130 flow of automated reasoning during the operation of a stored programme digital computer, 110 00:12:55,130 --> 00:13:01,520 which they were designing at that time. More about that later. 111 00:13:01,520 --> 00:13:11,150 And my point is phenomenological and applies. I will dare to say to the whole history of verbal communication that the media mediates. 112 00:13:11,150 --> 00:13:21,300 That tool shaped the thoughts and actions of the person using them and themselves embody what philosopher Davis Baird has called thing knowledge. 113 00:13:21,300 --> 00:13:32,330 The question for me is how the thing knowledge of the digital machine does this differently. 114 00:13:32,330 --> 00:13:38,540 The best way to prise open my topic is with the concept and practise of what is called modelling. 115 00:13:38,540 --> 00:13:43,310 That is the iterative manipulation of a software model. 116 00:13:43,310 --> 00:13:51,600 This is where computing, as we know it begins with a digital representation of whatever it is to be manipulated. 117 00:13:51,600 --> 00:13:58,710 There are two kinds of model, something in order to develop a model of something in order to develop one's understanding 118 00:13:58,710 --> 00:14:05,400 of it and a model for something imagined or known only from surviving evidence, 119 00:14:05,400 --> 00:14:10,380 either kind is unnecessarily simplified construct according to the machine's 120 00:14:10,380 --> 00:14:16,530 rigorous constraints of discrete all or nothing binary logic ones and zeros, 121 00:14:16,530 --> 00:14:25,380 as we say. What's important is that the model renders the real or imagined object of interest computationally tractable. 122 00:14:25,380 --> 00:14:33,450 That is, it represents the object in a completely explicit and absolutely consistent manner. 123 00:14:33,450 --> 00:14:42,990 These startlingly harsh axioms of digitisation, as I call them, have crucial implications I will return to in a moment. 124 00:14:42,990 --> 00:14:50,520 Modelling is thus not strictly mimetic of an object in the real world, nor an imagined one, 125 00:14:50,520 --> 00:14:56,580 because both require translation into binary terms and choice of what is to be included. 126 00:14:56,580 --> 00:15:02,220 Hence rigorous compromise in return for such sacrifices of truth. 127 00:15:02,220 --> 00:15:10,950 The model receives enormous combinatorial, manipulative power over the modelled object or idea. 128 00:15:10,950 --> 00:15:20,100 Thus, the essential tension of the digital trade off between memetic fidelity and computational manipulative bility. 129 00:15:20,100 --> 00:15:26,160 Now, let's first consider the analytic kind modelling of something more closely. 130 00:15:26,160 --> 00:15:33,360 It's illustrated here I will return to the second synthetic kind later for modelling 131 00:15:33,360 --> 00:15:40,170 of note in the slide the progressive circularity of represent of experiencing, 132 00:15:40,170 --> 00:15:49,350 simplifying, building, manipulating and comparing with improved understanding of the modelled object and of the resultant process. 133 00:15:49,350 --> 00:15:54,360 But consider what happens during the translation in the slide. 134 00:15:54,360 --> 00:16:02,490 Note what the modeller has to work with what he is constrained to leave out how much he is imagining. 135 00:16:02,490 --> 00:16:07,170 Let me give you three illustrative examples in the form of thought experiments. 136 00:16:07,170 --> 00:16:20,960 One. Art historical into literary. Consider these eight paintings on the story of Icarus painted from the 16th, the 21st century. 137 00:16:20,960 --> 00:16:26,000 The titles tell us they are about Icarus and in some cases, more than others. 138 00:16:26,000 --> 00:16:30,890 So the visual clues from a strictly computational point of view. 139 00:16:30,890 --> 00:16:39,860 The problem is deriving a rigorous set of criteria by which an algorithm could discover all these paintings without the help of their titles, 140 00:16:39,860 --> 00:16:50,150 or if it had those titles, how it could qualify their variations on the theme, as Marina Werner wrote in the latest London review of books. 141 00:16:50,150 --> 00:16:56,540 When the facts are missing, we tell stories. The machine works only from the facts. 142 00:16:56,540 --> 00:17:08,430 We tell the stories. Second. Are two chapters from Barbara Kingsolver is wonderful novel The Poison Wood Bible. 143 00:17:08,430 --> 00:17:16,860 Here we have a novel about an American family taken to what was then the Belgian Congo by their evangelical father, a fiery Baptist. 144 00:17:16,860 --> 00:17:27,470 Each chapter is narrated by a member of the family. The narrator's name is the title, except for the last chapter, which has no title. 145 00:17:27,470 --> 00:17:34,510 We can figure out with some work that the narrator is the dead child, Ruth May. 146 00:17:34,510 --> 00:17:46,650 Or rather, is the being to whom she reverted when she died the to who is nameless and all seeing encompassing all people born, unborn and dead. 147 00:17:46,650 --> 00:17:53,070 The problem is how to follow that inference algorithmically, not to the name Ruth Mae, 148 00:17:53,070 --> 00:17:59,760 which would be totally false, but to the Moon to whom Ruth may once instantiated. 149 00:17:59,760 --> 00:18:06,830 The problem is that the nameless narrator cannot be named. 150 00:18:06,830 --> 00:18:11,600 Next is this wonderful poem by Seamus Heaney. I won't read it. 151 00:18:11,600 --> 00:18:28,390 I'll let him do that. The railway children, when we climbed the slopes of the cutting, 152 00:18:28,390 --> 00:18:33,100 we were eye level with the white cops of the telegraph poles and the sizzling 153 00:18:33,100 --> 00:18:41,380 wires like lovely freehand they curved for miles east and miles west beyond us, 154 00:18:41,380 --> 00:18:51,340 sagging under their burden of swallows. We were small and thought we knew nothing worth knowing. 155 00:18:51,340 --> 00:19:01,300 We thought words travelled the wires in the shiny pouches of raindrops, each one seeded full with the light of the sky, 156 00:19:01,300 --> 00:19:14,380 the gleam of the lines and ourselves so infinitesimally scaled we could stream through the eye of a needle. 157 00:19:14,380 --> 00:19:20,700 Now, the problem I want to direct your attention to here is the last six words. 158 00:19:20,700 --> 00:19:29,120 Which, as you probably know immediately, is an allusion to the parable of the rich man in the Synoptic Gospels. 159 00:19:29,120 --> 00:19:37,080 But consider what it means to identify that illusion as opposed to leaving it be. 160 00:19:37,080 --> 00:19:44,540 Leaving it suspended to leaving it something, you know, but is not said. 161 00:19:44,540 --> 00:19:53,150 That's the difference I'm trying to get to in all three cases, we are left with a return to the drawing board and a number of questions. 162 00:19:53,150 --> 00:19:57,500 These spur us to ask first, how do we know what we know? 163 00:19:57,500 --> 00:20:08,190 Then to talk to a programmer or to ourselves as programmer even better about how a better response might be implemented. 164 00:20:08,190 --> 00:20:14,490 But let me return to the crucial implications of the requirement for computational tractability, as I've called it, 165 00:20:14,490 --> 00:20:22,170 the requirement that real world objects be rendered in a completely explicit and absolutely consistent manner. 166 00:20:22,170 --> 00:20:30,630 Clearly, however, much pain is entailed by fitting a poem or other work of art into such a pro crusty embed. 167 00:20:30,630 --> 00:20:39,690 We know how to do it. We know that the act of translation into discrete all or nothing terms illuminates by identifying the untranslatable. 168 00:20:39,690 --> 00:20:46,430 We know that once that translation is done, the machine has absolutely no problem crunching the bits. 169 00:20:46,430 --> 00:20:54,200 But how about the problem the interpreter has when he or she attempts to re translate the result back into human terms? 170 00:20:54,200 --> 00:20:57,070 That, to my mind, is the nub of the matter. 171 00:20:57,070 --> 00:21:06,740 Again, the negotiation for modelling of this is indeed the central and most difficult question for help with it. 172 00:21:06,740 --> 00:21:13,220 I turn to historian of science David Goodling's work on the ways in which an experiment bitter construed as, 173 00:21:13,220 --> 00:21:20,290 he says, new knowledge of the world from the behaviour of his experimental apparatus. 174 00:21:20,290 --> 00:21:26,820 Here's my version of his diagram in which the artefact of study. 175 00:21:26,820 --> 00:21:34,800 Is translated in a reductive fashion of absolute consistency and complete explicitness into a model in a programme, 176 00:21:34,800 --> 00:21:36,870 then some stuff happens in the black box. 177 00:21:36,870 --> 00:21:44,010 More about that black box in a moment and then is expanded back into the human context by means of these controls or, 178 00:21:44,010 --> 00:21:55,000 as he says, the flexible quasi linguistic messengers between the perceptual and the conceptual. 179 00:21:55,000 --> 00:22:03,060 Brilliant work. Gooding drew his theory of construal from a lifelong study of the 19th century scientist Michael 180 00:22:03,060 --> 00:22:10,250 Faraday's meticulously detailed laboratory notebooks attempting to recreate what Faraday did. 181 00:22:10,250 --> 00:22:15,320 Gooding focus specifically on that re translation from the controlled environment 182 00:22:15,320 --> 00:22:21,900 of the experimental situation to the human world of the experimenter. 183 00:22:21,900 --> 00:22:24,810 This took him into the phenomenological psychological, 184 00:22:24,810 --> 00:22:32,220 cognitive and sensuous processes of negotiation by which the experimenter draws on what he called construal, 185 00:22:32,220 --> 00:22:38,220 as I said, to form into provisional scientific knowledge. 186 00:22:38,220 --> 00:22:46,740 I'm strongly inclined to think that Gooding theory maps very well onto what happens when we engage in modelling of a textual corpus, 187 00:22:46,740 --> 00:22:52,680 a painting, a musical idea and so forth to make sense of them. 188 00:22:52,680 --> 00:23:01,830 But much more work is required of us. We need the equivalent of Faraday's laboratory notebooks for research done with computing in the humanities. 189 00:23:01,830 --> 00:23:11,280 We need to marshal the work of many decades on human computer interaction, especially in its in its attention to performance computing as theatre. 190 00:23:11,280 --> 00:23:17,010 To quote Brenda Laurel, we need to think about what it means to treat computational work in the humanities, 191 00:23:17,010 --> 00:23:25,980 analogously to experimental work in the laboratory sciences and so pay particular attention to the cognitive psychological aspects of work. 192 00:23:25,980 --> 00:23:34,500 By Gooding. Ryan T. Nancy in recession and others, many others phenomenology should come into play. 193 00:23:34,500 --> 00:23:43,340 If we do all that, we will be far better equipped to design a real digital critical edition. 194 00:23:43,340 --> 00:23:48,560 Allow me to put to one side my ignorance and naivete as an amateur in these fields. 195 00:23:48,560 --> 00:23:56,360 Literally a lover of what they do, but no specialist to insist on one thing that we never let slip from sight. 196 00:23:56,360 --> 00:24:04,640 Those axioms of digitisation, the absolute consistency and complete explicitness of digital representation. 197 00:24:04,640 --> 00:24:13,170 Some would say, and in fact do say that the genius of the digital is that renders digital superfluous. 198 00:24:13,170 --> 00:24:19,850 As it is when we listen to a digitally recorded music that some of us can't hear that it's digital. 199 00:24:19,850 --> 00:24:28,100 For research, I think Aiden Evans in his recent book, The Logic of the Digital is basically right that the discrete all or nothing quality of the 200 00:24:28,100 --> 00:24:35,300 digital runs all the way from hardware circuitry to the user interface into the resources we use. 201 00:24:35,300 --> 00:24:41,570 I conclude that to ignore the digital ity of the digital is to obscure the foil against which 202 00:24:41,570 --> 00:24:50,960 our reasoning struggles and is perhaps transformed when we do the kind of work I am describing. 203 00:24:50,960 --> 00:24:59,240 There's a very big question here that I haven't thought about nearly enough about to face in public directly. 204 00:24:59,240 --> 00:25:04,360 Namely, reasonings changes in prolonged exposure to the machine. 205 00:25:04,360 --> 00:25:10,330 But I will dare to guess that there are four aspects of computing's influence to consider. 206 00:25:10,330 --> 00:25:15,250 One is defined by the foil provided by those two axioms, as I just said. 207 00:25:15,250 --> 00:25:21,390 One has to do with the effects of Googling on the conventions of normal academic discourse. 208 00:25:21,390 --> 00:25:29,790 Richard Rorty, who writes well about this, has suggested the great American shift from metaphors of depth to metaphor is of breath, 209 00:25:29,790 --> 00:25:35,280 from probing one thing deeply to assembling and comparing many things. 210 00:25:35,280 --> 00:25:51,770 But I say no more about that here. The last two aspects the combinatorial and simulative powers of the machine take up the rest of my time here. 211 00:25:51,770 --> 00:26:03,140 OK, back to my adaptation of Goodwin's diagram, with emphasis added in the middle bit to shift the discussion from the reductive translation and the 212 00:26:03,140 --> 00:26:10,880 expansive translation via construal to that middle stage to the black box where the bits are crunched here, 213 00:26:10,880 --> 00:26:18,540 perhaps especially intellectual weed control. Clifford Geertz this term is needed. 214 00:26:18,540 --> 00:26:27,210 Many have said following Lady out of Lovelace's dictum that the machine can only do what it is told to do. 215 00:26:27,210 --> 00:26:36,120 Very, very reassuring that or in the anxious language of the mid 20th century, the computer is but a fast moron. 216 00:26:36,120 --> 00:26:46,020 In the 1950s, IBM made the slur doctrine to salve public fears of artificial intelligence, and then the mantra went viral. 217 00:26:46,020 --> 00:26:53,380 If you look into the literature of the period, it's everywhere. But such is not the machine we have, 218 00:26:53,380 --> 00:27:03,160 which in essential respects is the machine Herman Goldstein and John von Neumann addressed in their 1947 report on what programming it would entail. 219 00:27:03,160 --> 00:27:10,890 The first ever paper on programming. They pointed out that the difference in design, which makes the crucial difference, 220 00:27:10,890 --> 00:27:17,610 is the provision that allows a running programme to conditional on the outcome of previous operations. 221 00:27:17,610 --> 00:27:23,760 And more recently conditional on refine things in the outer world that it interacts with to deviate 222 00:27:23,760 --> 00:27:30,130 from the linear sequence of instructions or even to rewrite those instructions on the fly. 223 00:27:30,130 --> 00:27:38,700 Here again is Goldstone's and von Neumann's flow diagram, as they called it, illustrating the point. 224 00:27:38,700 --> 00:27:44,940 They explained, and I quote the words here, and I want you to to notice them. 225 00:27:44,940 --> 00:27:49,950 They explained that coding is not a static process of translation, 226 00:27:49,950 --> 00:27:57,510 but rather the technique of providing a dynamic background to control the automatic evolution of a meaning 227 00:27:57,510 --> 00:28:06,430 as the machine follows unspecified routes in unspecified ways in order to accomplish specified tasks. 228 00:28:06,430 --> 00:28:14,290 That's Herbert Simon's remark that the computer does what it has been told, but not in the sense usually intended. 229 00:28:14,290 --> 00:28:22,040 And Marvin Minsky's observation that the digital computer has given the word machine a new meaning. 230 00:28:22,040 --> 00:28:29,660 The high level of complications which result from this design, Goldstein and von Neumann went on to note, are not hypothetical or exceptional. 231 00:28:29,660 --> 00:28:35,030 They are indeed the norm. The power of the machine is essentially due to them. 232 00:28:35,030 --> 00:28:40,480 That is to the extensive combinatorial possibilities which they indicate. 233 00:28:40,480 --> 00:28:44,320 In essence, as von Neumann suggested, four years later, 234 00:28:44,320 --> 00:28:54,200 machines of the digital all or nothing type work by combining and recombining the data under given constraints until coherent patterns emerge. 235 00:28:54,200 --> 00:29:01,390 In a nutshell, then the value added at this stage is combinatorial. 236 00:29:01,390 --> 00:29:08,170 For an analogy, to make this somewhat more familiar. Take what happens in a research library. 237 00:29:08,170 --> 00:29:12,910 Which provides a large number of modular resources in a standard format so that a 238 00:29:12,910 --> 00:29:18,970 variety of readers with unforeseen purposes may combine and recombine them ad lib. 239 00:29:18,970 --> 00:29:26,290 Now we've had such a device with us, at least since the library of Asher Bonner Paula in the 7th century before the Christian era. 240 00:29:26,290 --> 00:29:36,190 If I'm not mistaken, on a larger scale, in more recent form, we see more or less the same with the web on a smaller scale within a single codex, 241 00:29:36,190 --> 00:29:41,260 particularly obvious when it is designed as a reference work, 242 00:29:41,260 --> 00:29:48,140 such as a critical addition built to foster recumbent material liberties and of course, the works of Voltaire. 243 00:29:48,140 --> 00:29:53,410 And it's come immediately to mind as fit corpora for such treatment. 244 00:29:53,410 --> 00:29:56,920 But my point is the familiarity of this way of working, 245 00:29:56,920 --> 00:30:04,990 though by now perhaps unfamiliar means of statistical tools for finding patterns in masses of data. 246 00:30:04,990 --> 00:30:09,820 Once again, distant reading is the most obvious example of work. 247 00:30:09,820 --> 00:30:12,550 Surprises from the process begin to emerge. 248 00:30:12,550 --> 00:30:22,450 Alan Turing suggested in 1950 by analogy with the critical mass of a nuclear reaction to a quantitative threshold of complexity. 249 00:30:22,450 --> 00:30:29,680 Note you add a little bit and add a little bit and add a little bit, and all of a sudden different things start to happen in digital humanities. 250 00:30:29,680 --> 00:30:35,380 We find cogent surprises from style, a metric analysis as well as from distant reading. 251 00:30:35,380 --> 00:30:42,310 For example, see the work of the Australian John Burroughs and the pamphlets of the Stanford Literary Lab. 252 00:30:42,310 --> 00:30:56,320 All 16 of them, I think there are no. Including the latest one by somebody and Moretti about finding patterns in large number of images. 253 00:30:56,320 --> 00:31:02,330 The last aspect of reasoning with the machine. 254 00:31:02,330 --> 00:31:11,390 That I will consider begins with synthetic modelling for modelling for is commonplace in engineering as a way of converging on an optimal design. 255 00:31:11,390 --> 00:31:16,070 For example, of an aeroplane wing in historical or archaeological research, 256 00:31:16,070 --> 00:31:22,220 its aim is to create a symbolic room of a phenomenon that no longer exists in whole or in part 257 00:31:22,220 --> 00:31:28,790 from whatever evidence is at hand and whatever reliable conjectures may still be possible. 258 00:31:28,790 --> 00:31:36,740 Examples are the Roman Forum and a greater challenge. The Theatre of Pompey in Rome, of which very little visible remains. 259 00:31:36,740 --> 00:31:43,120 These have been reconstructed visually by by such modelling. You can find all this on YouTube. 260 00:31:43,120 --> 00:31:50,080 Modelling for can also be used in the manner of a thought experiment to speculate about what might be or might have been. 261 00:31:50,080 --> 00:31:59,380 Historians don't like that very much, but it's useful in teaching, apparently modelling for blurs into simulation when the model is as it were, 262 00:31:59,380 --> 00:32:06,460 turned loose to see what can be learnt from it, from interacting with it and changing its parameters. 263 00:32:06,460 --> 00:32:12,960 Simulation is used to study phenomena that are difficult or impossible to observe. 264 00:32:12,960 --> 00:32:18,280 Or as the historian of the US Manhattan Project, the bomb project. 265 00:32:18,280 --> 00:32:28,180 Wrote. Phenomena which are too far from the course of ordinary terrestrial experience to be grasped immediately or easily. 266 00:32:28,180 --> 00:32:33,670 The physicists of Los Alamos reached beyond their terrestrial experience by simulating the random 267 00:32:33,670 --> 00:32:40,390 interactions of neutrons in a nuclear chain reaction using the so-called Monte Carlo technique. 268 00:32:40,390 --> 00:32:45,250 Peter Gillison has written very illuminating Leigh on the subject closer to our own interests. 269 00:32:45,250 --> 00:32:49,210 Linguists have used simulation to study the migration of dialects. 270 00:32:49,210 --> 00:32:55,240 Naval historians to approximate the course of the battle at sea. 271 00:32:55,240 --> 00:33:03,100 In the physical sciences, including climatology simulation was just about evil with the invention of digital computing machinery. 272 00:33:03,100 --> 00:33:08,880 Indeed, the sciences were a driving force in its early development. 273 00:33:08,880 --> 00:33:13,890 Work in economics and other social sciences followed soon after. 274 00:33:13,890 --> 00:33:20,100 Most of us have forgotten that there is a rich and valuable history of simulation in the creative arts from the 275 00:33:20,100 --> 00:33:28,410 nineteen sixties and seventies of possible worlds and of works of art co-created interaction with the viewer. 276 00:33:28,410 --> 00:33:34,320 There is, in other words, a long and complex tradition of modelling of the more imaginative, 277 00:33:34,320 --> 00:33:44,360 less mimetic kind that offers us lessons and risk-taking and its rewards from which we have very much to learn. 278 00:33:44,360 --> 00:33:54,350 I want to conclude by giving you some idea of the limb out on which I now invite you to claim for a view of new things. 279 00:33:54,350 --> 00:34:01,730 The philosopher Paul Humphreys, University of Virginia, argues that in going out on that limb with computing beyond what we can do, 280 00:34:01,730 --> 00:34:08,090 otherwise, scientific epistemology ceases to be human epistemology. 281 00:34:08,090 --> 00:34:10,070 I think that David Gooding is right, 282 00:34:10,070 --> 00:34:20,310 that the humanness of our knowledge is a matter of what we do with what we learn in extent so from what we could otherwise not reach. 283 00:34:20,310 --> 00:34:23,460 Before I conclude with an example illustrating what I mean with that, 284 00:34:23,460 --> 00:34:31,710 with keen awareness of how nervous we are apt to get about venturing out or up beyond the ground, we take to be solid. 285 00:34:31,710 --> 00:34:37,210 Allow me to recommend to bracing touchstones. First. 286 00:34:37,210 --> 00:34:49,180 American essayist Elaine Scaries 1992 article The Made Up and the Made Reel Second Part B of Canadian philosopher Ian Hacking's 1983 book, 287 00:34:49,180 --> 00:34:58,090 representing an intervening. Experts in these areas will have more recent items to recommend I trust, but these two, 288 00:34:58,090 --> 00:35:08,190 as well as several others that I absolutely forbid myself to mention, mark important moments of light on my subject. 289 00:35:08,190 --> 00:35:08,760 Elsewhere, 290 00:35:08,760 --> 00:35:16,890 I have argued that it is far more useful and true to the practise of simulation that is to the use of the machine to reach for knowledge too 291 00:35:16,890 --> 00:35:25,600 far from the course of ordinary terrestrial experience to regard the machine not merely as a tool for pushing back the fence of the law. 292 00:35:25,600 --> 00:35:31,560 As Jacob Pronouncer characterised science nor merely as a tool for adventurous play, 293 00:35:31,560 --> 00:35:40,950 but as a processes for the imagination, the range is a long, lean, scary spectrum from the making up to the making real. 294 00:35:40,950 --> 00:35:48,000 I have suggested that at least from the perspective of the disciplines of baking simulation is the essential genius of the machine. 295 00:35:48,000 --> 00:35:52,950 And so what we must understand about it and learn to employ. 296 00:35:52,950 --> 00:35:58,450 I have suggested the simulation as a tool for restructuring our experiences of the world. 297 00:35:58,450 --> 00:36:09,340 That, as anthropologist Yaron Mimica says of the combinatorial ethanol mathematics of the way people in Indonesia, it is Mithal poetic. 298 00:36:09,340 --> 00:36:16,300 Other recent work in ethanol mathematics and in the history of combinatorics suggests that cosmological world making by 299 00:36:16,300 --> 00:36:23,560 combining and recombining units of experience such as the digits we are born with is as close to a universal language. 300 00:36:23,560 --> 00:36:36,560 After Babbel, as we are likely to get. Labelmates is calculated us and our own combinatorics belong in this worldwide tradition. 301 00:36:36,560 --> 00:36:38,390 Fascinating, isn't it, 302 00:36:38,390 --> 00:36:48,220 that the device from which we have so often expected closure turns out to be so paradoxically a tool that does exactly the opposite? 303 00:36:48,220 --> 00:36:54,850 Historian of biology, Evelyn Fox Keller has cautioned us, read everything she's written, she's wonderful. 304 00:36:54,850 --> 00:36:59,890 Noting that simulation changes sometimes radically from discipline to discipline, 305 00:36:59,890 --> 00:37:09,790 I have argued for its continuity across disciplines so as to project it where it is not had much hit, where it is not yet made much headway. 306 00:37:09,790 --> 00:37:17,620 What then might it look like in the disciplines of the humanities for a suggestive answer? 307 00:37:17,620 --> 00:37:26,470 The best I know I turned to John Wall's Virtual Poles Cross project, which has led already to the larger virtual St Paul's. 308 00:37:26,470 --> 00:37:32,230 It is an auditory simulation of John Dunne's Gunpowder Day sermon at Paul's cross, 309 00:37:32,230 --> 00:37:40,270 as it was in London on 5th November 16 22 with supporting visualisations, which that is one. 310 00:37:40,270 --> 00:37:49,720 As you will know, both the mediaeval St Paul's Cathedral and the Poles Cross Preaching Station, which you see there on the right within its grounds, 311 00:37:49,720 --> 00:37:55,860 were destroyed together with the surrounding buildings and the great fire of London in 60 and six. 312 00:37:55,860 --> 00:38:01,980 So in simulating the acoustics there, there is much room for conjecture. 313 00:38:01,980 --> 00:38:09,210 Walls aim he's a brave man is to explore early modern preaching through performance by simulating 314 00:38:09,210 --> 00:38:16,650 all of what Dan's congregation would could reasonably expected to have heard on that day. 315 00:38:16,650 --> 00:38:22,620 While in a number of other early modernists argue that the sermons of the time were their performances. 316 00:38:22,620 --> 00:38:31,830 The text we have merely traces of them consider an argument by analogy. 317 00:38:31,830 --> 00:38:39,000 Would we mistake the snippet in the upper left from the score of the Goldberg variations for box music? 318 00:38:39,000 --> 00:38:44,910 The bid at the lower right from the script of Tennessee Williams A Streetcar named Desire for the play. 319 00:38:44,910 --> 00:38:52,980 Of course not. We would know immediately that they are instructions for interpretive performances of the music and the drama, respectively. 320 00:38:52,980 --> 00:39:03,010 Just as now, I I hope you will think of programming code such as this as instructions for a different sort of interpretive performance. 321 00:39:03,010 --> 00:39:10,780 But we are so inured to silent reading that we are likely to regard this kind of thing as the primary object. 322 00:39:10,780 --> 00:39:17,770 And so by extension, to overlook that for the sermon in pols words from his letter to the Romans. 323 00:39:17,770 --> 00:39:29,380 Faith comes by hearing. Social cohesion and political unity come also as in the case of this sermon, polocrosse sermons were composed. 324 00:39:29,380 --> 00:39:33,250 It's temporary from notes by a preacher who faced a congregation. 325 00:39:33,250 --> 00:39:39,430 This is wall talking face to congregation gathered in open air and surrounded by a host of distractions, 326 00:39:39,430 --> 00:39:46,330 including the birds, the dogs, the horses, the bells and each other. 327 00:39:46,330 --> 00:39:53,020 Dramatic engagement with the congregation was essential to their success. 328 00:39:53,020 --> 00:39:59,690 What do we need as scholars in order to get closer to the sermon as it actually was? 329 00:39:59,690 --> 00:40:09,960 Near contemporary evidence have done sermon was confined to that first printed edition of sixteen forty nine on the left. 330 00:40:09,960 --> 00:40:21,540 Until Jeanne Shami identified British Library manuscript Royal 17B 20 in 1995 as a scribble presentation copy corrected in Dunn's own hand. 331 00:40:21,540 --> 00:40:32,280 This is very likely the manuscript that was produced at the request of James, the first from Dunn's, notes hours, days or weeks after the event. 332 00:40:32,280 --> 00:40:40,230 He preached from notes, not from a text, as I am twenty seven years passed before the text was printed. 333 00:40:40,230 --> 00:40:45,870 The great fire of destroyed almost everything, but not quite everything relevant to the reconstruction. 334 00:40:45,870 --> 00:40:51,070 Just 17 years after that. So what is going on? 335 00:40:51,070 --> 00:41:01,160 The project? While his written is about what we are doing when we believe we have discovered from experience with a digital environment. 336 00:41:01,160 --> 00:41:06,450 Things about past events that are not documented by traditional sources. 337 00:41:06,450 --> 00:41:13,380 Indeed, what are we doing? This is the brink of the new or is it so very new? 338 00:41:13,380 --> 00:41:19,680 One may wish to retreat to safer ground, but I wonder just how safe that ground is when one looks closely, 339 00:41:19,680 --> 00:41:23,870 how much we build inferential step by inferential step. 340 00:41:23,870 --> 00:41:32,210 And is safety our goal positive knowledge of the sort Latter-Day Positive this quest for is ephemeral, the history, 341 00:41:32,210 --> 00:41:39,020 philosophy, sociology and psychology of the sciences from Thomas Kroon onwards furnish a good corrective. 342 00:41:39,020 --> 00:41:50,660 Again, I refer you to Ian Hackings writings. We must, of course, be scrupulously careful and cautious for imagination as a wild beast. 343 00:41:50,660 --> 00:42:00,850 My Hope for Wolves project. His wonderful project is that someday soon it can become a dynamic simulation in which you 344 00:42:00,850 --> 00:42:06,940 and I can twiddled the knobs and see what happens if this or that conjecture is altered. 345 00:42:06,940 --> 00:42:13,870 What if that building were there? What if the building material were something else and so on and so forth? 346 00:42:13,870 --> 00:42:21,100 Surreal Lootera. Seriously, to play and to play imaginatively is what our machine is all about. 347 00:42:21,100 --> 00:42:28,390 Only Wall and his collaborators at North Carolina State University can do that now, and only quite slowly. 348 00:42:28,390 --> 00:42:36,880 And I very much hope that they, as they work, are keeping laboratory notebooks so that we can begin to follow in David Goodwin's footsteps. 349 00:42:36,880 --> 00:42:47,740 To do that, we need a much wider, more institutionally committed gathering of the disciplinary tribes here that disciplinary tribes in my book, 350 00:42:47,740 --> 00:42:56,640 that's what digital humanities is for. Thank you very much.