In his book Humanities Computing, in a chapter fittingly titled "Discipline," Willard McCarty attempts to define the discipline of Humanities Computing by understanding what is happening within the discipline. According to McCarty it is through this understanding that "we may get to the disciplinary conditions from which specific methods arise as desire or need direct" (McCarty 2005, 115). This suggests that one way to understand the beginnings of a discipline are to look not at the founding people, but at the desires and needs articulated at the time. This paper will therefore look at how an agenda was set for humanities computing in English Canada around a particular moment of emergence, in this case the emergence of a scholarly society, the Ontario Consortium for Computing in the Humanities, which evolved into the national Consortium for Computers in the Humanities (COCH/COSH). To do that we will outline the series of events in the late 1980s starting with the formation of the Centre for Computing in the Humanities at the University of Toronto (CCH). We will conclude by turning to issues related to documentary evidence and methodology when doing the history of computing.
Defining the discipline
The question of whether humanities computing (or the digital humanities) is a discipline has been discussed to the point where few are interested any longer. The University of Virginia held a year-long seminar on the subject titled "Is Humanities Computing a Discipline?" that exhausted the issue (The seminar papers are all available at http://www.iath.virginia.edu/hcs/index.html. Some of these were published online in the Jahrbuch für Computerphilologie (Issue 4, 2002) at http://computerphilologie.uni-muenchen.de/jahrbuch/jb4-content.html). Melissa Terras (2006) followed up with an in depth analysis of humanities computing through its research, community and curriculum.
The field, whether you consider it a discipline or not, has many of the structures of a discipline. It has scholarly associations like the Association for Computing in the Humanities, the Association for Literary and Linguistic Computing, and the Canadian Society for Digital Humanities with conferences. These associations have journals like Literary and Linguistic Computing, Digital Studies, and Digital Humanities Quarterly, to name just the English journals. There are academic programs that perpetuate the "discipline" by training new "disciples" like the Humanities Computing program at the University of Alberta and the Doctorat en Informatique Cognitive at the Université du Québec à Montréal (a wiki with information about digital humanities programmes, labs, centres, institutes,projects, journals, professional societies, etc. can be found at http://tapor.ualberta.ca/taporwiki/index.php/The_Academic_Capacity_of_the_Digital_Humanities_in_Canada). Book series are flourishing and Stanley Fish has even discussed the digital humanities as the new thing in three of his Opinionator blog entries for the New York Times (Fish even has a Digital Humanities tag that you can use to see the three blog entries, http://opinionator.blogs.nytimes.com/tag/digital-humanities/). If it walks like a discipline and quacks like a discipline, then it probably is one.
The more interesting questions to ask now, before the record is lost, are historical questions like, how did humanities computing emerge and how did it develop the structures of a discipline? This, however, raises issues about how to do the history of a field like humanities computing. One way is to look at what people say about the history in publications. One can look at how Father Busa describes his Index Thomisticus project (Busa 1980). We believe, however, that personal memoirs of projects are only part of the picture. If it is disciplinarity that we want to understand then we need to understand the emergence of the structures like centres, associations, journals, instructional programmes and conferences. These structures are often poorly documented in the published record as they do not constitute research so much as administrative advances. The very centres, associations, and journals that provide the framework for research often neglect to document their enframing. The structures are supposed to disappear before the research done, which makes it hard to understand how they might have influenced the research done.
It is particularly difficult to recapitulate the history of humanities computing centres, which is a pity as we believe that centres were central to the structural history of the field. Humanities Computing centres often began as service units that assisted researchers with technical challenges, rarely initiating intellectual inquiry or publishing about themselves. As these units grew, and as the discipline of digital humanities coalesced around topics such as text analysis and language learning, the initiative for new projects shifted with this disciplinary self-awareness. It is in understanding these moments of creation that the identity of the humanities computing discipline can best be understood. For this reason history looks back to Herotodus and Leopold von Ranke the supposed fathers of history and historiography, respectively. To what beginnings should computing humanists look back? Further, how can we do the historiography of humanities computing, and from which disciplinary histories may we draw parallels?
Certainly the historiography of history offers methodological starting points. Historians have struggled with the nature of their work for centuries, usually placing the importance of history higher than most disciplines. Thomas Carlyle, for example, proclaimed the immortality of history: "some nations have prophecy, some have not: but of all mankind, there is no tribe so rude that it has not attempted History, though several have not arithmetic to count Five" (Breisach 2007, 1). The thousands of years of disciplinary navel gazing, however, may not provide parallels useful enough to our field, which looks back a few decades and not multiple centuries. And while humanities computing may in some instances contribute insights to such traditions as economic history, social history, Marxist history, etc., Capital-H History may be too unwieldy, too cumbersome, and simply too long in the tooth to provide effective insight into the birth of our discipline. In addition, the strong methodological underpinnings that guide humanities computing inquiry stand in stark contrast to the general methodological blindness that shapes much of the traditional discipline of history. We should look elsewhere for inspiration.
The most interesting of these "elsewheres" may best be located in offshoots of the history of History. The emergence of statistical methodologies in general and quantitative history in particular seems a promising place to start. Certainly, there are commonalities of utility that both disciplines shared in their formative years prior to the emergence of their own intellectual investigations. Humanities computing, however, cannot hang its methodological hat with any particular mathematical certainty. Quantitative methods, as fraught with qualitative assessment and judgments as any discipline, can at least fall back on the law of large numbers. This fundamental theorem asserts, in part, that a sufficiently large sample, randomly selected, will produce stable, predictable results, and it serves as the bedrock for most of the statistical tools that social scientist of a particular bent employ. We have no such solidity, other than simply co-opting aspects of that methodology ourselves. Perhaps a similar assertion may be made about historical geography, although again, there are parallels that are worth investigating.
One offshoot of history, however, seems a particularly promising garden path to wander. The Annales school, founded (for lack of a better word) in the 1920s, promoted a new vision for the study of history: an analytical, problem-based exploration rather than a narrative of events; an exploration of a range of human experiences rather than a political teleology; and most importantly, an openness to the methodologies and affordances of any discipline willing to play along, rather than a narrow disciplinary view. In the words of Peter Burke, the founders of the Annalistes were "small, radical, and subversive, fighting a guerrilla actions against traditional history, political history, and the history of events" (Burke 1990, 3).
While such lines of inquiry prompt fascinating introspection, their very nature also demands a deliberative and considered methodology. The processes that shaped and formed the development of the humanities computing discipline were shaped and formed by members still active in our community. When our colleagues become the object of study, rather than fellow academics exploring with us, research obligations may require greater care. As Smart, Feldman, and Ethington noted in their study of academic disciplines, "knowledge of why people select disparate disciplinary settings and how they interact with others in these settings is essential to understanding the fundamental issues of interest to higher education researchers and policy makers - issues such as persistence, satisfaction, and success" (Smart, Feldman, and Ethington 2000, xi).
What emerges from this observation is the self-reflecting nature of the discipline formation, namely that the choice of vocation is an expression of the personality. In studying the emergence of our discipline, we end up also studying the personalities of those who gave form to the formless. We study those who gave and may still be giving structure to the field.
The earliest efforts in the field have been well documented if not mythologised. In "The History of Humanities Computing" (Schreibman et. al. 2004) Susan Hockey outlines the evolution of humanities computing beginning with the work of Father Roberto Busa and the development of various concordances through the establishment of conferences and associations into the early 1970s (Father Busa is traditionally seen as the father of humanities computing, but his priestly paternity has not seriously been questioned. Others were also active in late 1940s and 1950s, like Andrew Booth who worked on machine translation (and is considered a pioneer in that field.) He began at Birbeck College at the University of London, coming to Canada in 1962 (Rockwell 2007). Hockey also discusses the establishment of centres such as the Centre for Literary and Linguistic Computing in Cambridge, which was started in 1963; the first journals such as Computing in the Humanities, which was started by Joseph Raben in 1966; and the founding of associations such as the Association for Literary and Linguistic Computing in 1973.
The Canadian structural history is harder to gather, which is exactly the point of this paper – to reconsider the history of the field in Canada and identify what needs to be gathered. Some key early moments include the publication in 1966 by the University of Toronto Press of the manual for PRORA, a lemmatising concordance system designed for use in the romance languages (Glickman and Stallman 1966), reviewed in the first issue of Computing in the Humanities. In 1969, the Centre for Medieval Studies at University of Toronto launched the Dictionary of Old English (Leyerle 1971). Not long after, in 1973, researchers at the Université du Québec à Montréal implemented SATO (Systeme d’analyse des textes par ordinateur), a textual data processing system for literary and social sciences (Meunier et. al. 1976), and soon after created the Centre d’ATO. In 1977, Université de Montréal and University of Waterloo organised the first major international conference in Canada and the proceedings were published as an edited book (Lusignan and North 1977). This was the third International Conference on Computing in the Humanities following a series of conferences in the UK and the USA. Jumping ahead a decade, in 1987 Willard McCarty, then at the Centre for Computing in the Humanities (CCH) at the University of Toronto, started the Humanist discussion list, which is still going and which he still moderates (the archives of Humanist are at http://www.digitalhumanities.org/humanist/. You can see the original message welcoming and announcing the group at http://www.digitalhumanities.org/humanist/Archives/Virginia/v01/8705.1324.txt. You can search Humanist using Voyant at http://voyant-tools.org/?corpus=humanist). And in 1989 the Association for Computing in the Humanities and the Association for Literary and Linguistic Computing held their first joint conference at the University of Toronto, an event at which both Northrop Frye and Ted Nelson talked.
One of the most important structures for the emergence of humanities computing in Canada was the national scholarly association now called the Canadian Society for Digital Humanities / Société Canadienne des Humanites Numériques (CSDH/SCHN), which is the publisher of Digital Studies / Le champ numérique. The Canadian Society was, until the Annual General Meeting at the 2012 Congress, called just the Society for Digital Humanities / Société pour l'étude des médias interactifs (SDH/SEMI), which in turn used to be called COCH/COSH or the Consortium for Computers in the Humanities / Consortium pour ordinateurs en sciences humaines. COCH/COSH emerged in 1986 according to Ian Lancashire, its first President, out of the Ontario Consortium for Computing in the Humanities (OCCH) (Lancashire 2002).
The shift from a "consortium" to a "society" in 2005-6 is important to the structural history of the association. OCCH and COCH/COSH were both consortia of centres. The governance of these consortia was based on there being representatives from each participating centre. This can still be seen in the constitution of CSDH/SCHN, which has language about a Board of Directors where "Each member institution will be represented by one person chosen by that institution and referred to as the official representative of the member institution" (The constitution of the CSDH/SCHN as of June 2012 had not been updated with the name change from SDH/SEMI to CSDH/SCHN. It also has archaic language from when SDH/SEMI was a Consortium. See http://sdh-semi.org/about/constitution/). In effect, our national society, like the Ontario provincial one before it, was built by the directors of centres as a way to bring together university humanities computing centres (and people associated with them), not around a model of broad academic participation. This, if anything, indicates the importance of centres politically and their importance to building community through the 1980s. When did it change?
In the mid 1990s under the leadership of Ian Lancashire, COCH/COSH joined the Canadian Federation of Humanities and Social Sciences in order to have annual meetings as part of the Congress of the Humanities and Social Sciences. This was the beginning of a shift to a "society." The formal shift happened, as far as we can tell between the 2005 Congress and the 2006 Congress under the leadership of English President Ray Siemens and French President Christian Vandendorpe. If you search the Wayback Machine for “coch-cosh.ca”, the domain for the COCH/COSH web site, you find that site changes from being about COCH/COSH to being about SDH/SEMI between the crawl of 13 October 2005 and 10 December 2005. This probably reflects the decision of the executive and society at the AGM of 2005 to change the name and operation of the organisation. Searching the Humanist archives one also finds that the 2006 Congress conference is the first for which SDH/SEMI is the organisational name (we should add that author Geoffrey Rockwell was a Director of COCH/COSH and then Vice President of COCH/COSH and SDH/SEMI during this time).
Let us look more closely at the formation of COCH/COSH. This organisation, as mentioned above, evolved quickly out of an Ontario organisisation, OCCH. OCCH in turn was set up by Ian Lancashire of the University of Toronto and Elaine Nardocchio of McMaster University, both instrumental in setting up local university centres. Thus the national organisisation seems to have evolved out of a provincial one, which in turn came out of the setting up of university centres. This desire to bring together like-minded humanities computing scholars formally into some sort of organisisation began in the early 1980s and seems to have quickly leapt from the local to the national.
Looking locally, in the case of the University of Toronto Centre for Computing in the Humanities, we have documentation around the formation of the centre and its report from 1990 on "Humanities Computing: The CCH Toronto-IBM Canada Co-operative." The CCH was formed in 1986 with significant support from IBM Canada and Ian Lancashire kept the key documents around the CCH IBM Co-operative. In the 1990 report Lancashire described the efforts in Toronto in words straightforward in intent but ambitious in scope: "The goal of the co-operative was transferring technology to improve the productivity of the students and teachers, and the means to do so was the CCH. Its mandate was to place services, documentation and electronic communications in the public domain for all levels of education, business, research centres, and government" (Lancashire 1990, ix). One can see the importance of centres in the 1980s. Simply getting the technology into the hands of students and faculty was a key goal as otherwise most would not have any access whatsoever. Providing access in the form of labs and other services in turn would make it possible to use computing tools in their teaching and research. This would ultimately, in the words of Lancashire, "give our colleagues and students in the humanities the confidence that they could contribute, side-by-side with business, in the same global market" (Lancashire 1990, 10).
The Toronto Centre for Computing in the Humanities did not spring full formed. Lancashire in the 1985 "Letter from Toronto" published in Computers and the Humanities describes the history from the publication of the PRORA manual in 1966 up to 1986 when the university creates a formal university organisation to coordinate humanities computing efforts. He opens the letter, "Early in the summer of 1983 the University of Toronto took two steps in support of humanities computing. University President Jim Ham, an engineer, dedicated private funds within the university to set up semi- automatic editorial, typesetting and database computer facilities at two long-term research and publishing projects supported by three organizations: the Social Sciences and Humanities Research Council of Canada, the Dictionary of Old English, and Records of Early English Drama. In a complementary act of support, Professor David Nowlan, Vice-President (Research and Government Relations), established a central Humanities Support Group staffed by a fulltime programmer within University of Toronto Computing Services (UTCS). This office is a resource for individual researchers with computing needs as well as for university projects with independent computing facilities" (Lancashire 1985, 251). Form the 1960s until 1983 the history of humanities computing at the University of Toronto was that of individual projects that were not coordinated, according to Lancashire--though some of those projects, such as the Dictionary of Old English, were substantial (with significant facilities, funding, and staff). What changed in 1983 was the formation of institutional structures that went beyond particular projects. On the one hand, the University President set up publishing facilities at two projects, presumably for use by others too. More importantly the Vice-President set up a Humanities Support Group and gave them staff support. This group seems to have helped bring individuals and research projects together so that they could articulate their common needs and lobby for better support in the form of a Centre.
By 1983 there was an "Advisory Committee for the IBM/U of T Cooperative Agreement" which would indicate that IBM had shown some interest in supporting something. What emerges from these discussions is the Centre for Computing in the Humanities, which formally started in 1986. In 1989 CCH organises the first joint ACH/ICCH conference in Toronto and in 1990 we have Lancashire’s report on the three-year agreement between IBM and the University of Toronto. What is interesting is how the needs of humanists were articulated back then. The 1983 "Minutes of the Meeting of the Advisory Committee for the IBM/U of T Cooperative Agreement" lists the following "Needs of Researchers in the Humanities":
- data collection and communication links with Oxford and U.S. Universities, that have established databases of text,
- data base packages that would assist researchers to organize and analyze text,
- ability to handle odd characters and symbols on the terminal keyboard, screen and printer, …
- printing of high quality documents for publication,
- various software packages, e.g. one that can parse English,
- improvement of the interface between the researcher and the microcomputer (Minutes, 1).
- history of ideas and literary criticism (levels 1-3)
- text explication (1-4)
- historical editions, indexes, lexicography biographies, bibliographies, "literary" history (1-5)
- stylistics, authorship studies, language studies
- linguistics and cognitive psychology (Minutes, Appendix, 3)
- Provided instructional technology support for courses,
- Organized conferences and workshops,
- Provided network communication connects through NetNorth to the rest of the world,
- Published newsletters, a yearbook, and the Humanist discussion list, and
- "Provided substantial stimulus and help to create first a provincial and then a national consortium of over 22 universities in Canada to develop resources in for humanities computing" (Lancashire 1990, ix-x)
- A democratic consortium was preferable to one large national centre that would probably be located at one university. Perhaps colleagues outside of the University of Toronto were afraid of being having to always work through Toronto or Montreal.
- A consortium, like the coordination within the University of Toronto, might make it easier to get funding as the ATO Centre example demonstrated. The Report mentions an earlier attempt to create a Canadian National Centre in order to get funds from SSRHC. It was this centralised model that Nardocchio reports people objecting to in favor of a Consortium.
- Elsewhere in the summary Nardocchio also reports that respondents wanted more contact with others, directories, newsletters, lectures, courses, and "pools of resource people" (Nardocchio 1986, 2). Humanities computing was lonely and many wanted better communication.
- I suspect the organisational form identified would also have the side effect of raising the profile of the institutional representatives and identifying individuals as responsible for local leadership. Being a Director of the Consortium would give you a voice that other organisational forms would not, but it also would make you responsible. Thus the people involved setting up OCCH and then COCH/COSH might have profile reasons for wanting a national organisation.
In sum, some themes stand out in the documents we have gathered that document the triple organisational founding of CCH, OCCH and then COCH/COSH:
- Researchers at the time felt they needed access to equipment we now take for granted like microcomputers, scanners and software.
- Researchers felt that the best way to get access was to form committees to represent their needs.
- As committees successfully lobbied university administrations, those administrations formed centres or other types of central organisations that could house resources. Centres emerge logically from the tensions between a group of researchers with needs and the university that does not want to duplicate expensive resources.
- The first organisations that spanned universities in English speaking Canada grew out of leadership in Southern Ontario, notably Lancashire and Nardocchio at Toronto and McMaster.
- These first organisations were developed as consortia of universities, not as a society of members. It was amazing that OCCH could actually get most of Ontario's universities to formally sign on.
- The centres, organisations, and principals were significantly concerned with securing funding from SSHRC and the legitimisation that would offer. Many of the early activities of OCCH seem aimed at demonstrating humanities computing as a credible field with significant penetration. When OCCH turned quickly into the national COCH/COSH, these community formation activities continued at a national level.
We therefore see what we believe is a common feature of institutions during the fluid formative years: they engage in activities that present the organisation as an established and broad community in order to get the support and membership such that they can actually become what they say they are. In the case of OCCH/COCH/COSH this involved developing directories of people working in the field and surveying people in order to show the need for an organisation.
What remains clear in this study is the importance of unpublished administrative documents. Of the key documents used to trace this history only the "Letter from Toronto" was published. The crux was the examination of minutes and reports. For this reason we want to conclude by discussing the nature of evidence needed to understand the disciplinary history of the digital humanities and what has to be done to secure that evidence.
If we want to be able to trace the history of computing in the humanities we need to find and archive administrative documents. Computing has always been expensive and therefore researchers needed to organise, lobby, and form organisations to manage resources. The documents that record these activities are not trivial to understanding how computing was conceived and actually deployed. We hope we have shown how the "before" and "after" documents of the CCH can be used to leverage a picture of an important moment in the formation of the field. The CCH made possible many of the publications and courses that communicated and trained us, but it, like any other organisation, was structured in certain ways.
The challenge now, before the materials are lost, is to gather and properly archive such administrative documents. The Histories and Archives Project at the University of Alberta has begun to do that. The impetus for the project began with the discovery of boxes of documents that literally fell into our hands. In 2008, Geoffrey Rockwell rescued from the garbage boxes of materials gathered and preserved over the years by Terry Butler. These documents were further augmented with a similar collection belonging to Ian Lancashire at the University of Toronto. In an effort to compliment this collection team members began soliciting related materials around humanities computing activities at the University of Alberta as a number of key people are still around.
The growth of our collection pushed us to think about how to preserve and provide access to what we saw as a treasure. It was decided that a digital archive is particularly fitting for the subject material, and we therefore have developed a Histories of Humanities Computing collection with the University Education and Research Archive (ERA), a digital repository built on Fedora (ERA is available at https://era.library.ualberta.ca/, and our Collection at http://hdl.handle.net/10402/era.17176, though as of writing most of the materials are embargoed for reasons of copyright and privacy. Documentation of the decisions that went into the archive can be found at our wiki, http://circa.cs.ualberta.ca/index.php/CIRCA:Histories_and_Archives_Project). Over the last year the team has researched the various components of building a digital archive that would suit our specific needs. Significant back-end work goes into creating and building the archive, and it is important to understand the breadth of considerations to be mindful of when building, implementing, and maintaining the digital archive (Cohen and Rosenzweig's text, Digital History: A Guide to Gathering, Preserving, and Presenting the Past on the Web is a useful starting point).
One of the key issues that has arisen throughout this project is regarding the nature of the collected documents. As mentioned before, many of them were once part of a personal collection comprised of not only published newsletters but of Minutes of Meetings and personal correspondence - the potential copyright and privacy issues are evident. In addition, the majority of information we are currently dealing with has a focus on English Canada because of the sources. Although we have come across a few references that mention contemporaneous events in Quebec, such as the establishment of the Centre d'Analyse de Textes par Ordinateur (Centre d’ATO) at the Université du Québec à Montréal, a holistic picture of the emergence of Humanities Computing in Canada will take collaboration with other archives.
We end with a request. First, buried in banker’s boxes, archived in email folders, stuffed into filing cabinets, saved onto floppy disks, and written on paper are documents and correspondence that detail the history of our discipline. Much of this may seem like the detritus of administration. We hope, however, to have shown how important these materials can be. Help us save them. If you have such documents donate them to your university archive or, if they won’t take them, send them to us. Even better, digitise them and send us the digital copies to archive in our Histories of Humanities Computing collection.
Breisach, E. 2007. Historiography: Ancient, Medieval, and Modern. Third edition. Chicago: University of Chicago Press. Print.
Busa, R. 1980. "The Annals of Humanities Computing: The Index Thomisticus." Computers and the Humanities. 14:2. 83-90. Print.
Burke, P. 1990. The French Historical Revolution: The Annales School, 1929-1989. Stanford: Stanford University Press. Print.
Cohen, D. and R. Rosenzweig. 2005. Digital History: A Guide to Gathering, Preserving, and Presenting the Past on the Web. Philadelphia: University of Pennsylvania Press. Print.
Glickman, R. J. and G. J. Staalman. 1966. Manual for the Printing of Literary Texts and Concordances by Computer. Toronto: University of Toronto Press. Print.
Lancashire, I. 2002. "Humanities Computing and Research Innovation." (paper). COCH/COSH 2002. http://web.viu.ca/siemensr/C-C/2002/Lancashire.htm.
Lancashire, I. 1990. "Humanities Computing: The CCH Toronto-IBM Canada Co-operative." Report published by the Centre for Computing in the Humanities, University of Toronto. From the papers of Ian Lancashire.
Lancashire, I. 1985. "Letter from Toronto." Computers and the Humanities, 19:4. 251-253. Print.
Lancashire, I. 1983. Minutes of the Meeting of the Advisory Committee for the IBM/U of T Cooperative Agreement. From the papers of Ian Lancashire.
Leyerle, John. 1971. "The Dictionary of Old English: A Progress Report 1." Computers and the Humanities. 5:5. 279-283.
Lusignan, S. and J. S. North, eds. 1977. Computing in the Humanities: Proceedings of the Third International Conference on Computing in the Humanities. Waterloo, Ontario: University of Waterloo Press. Print.
McCarty, W. 2005. Humanities Computing. Basingstoke: Palgrave MacMillan. Print.
Meunier, J., S. Rolland, and F. Daoust. 1976. "A System for Text and Content Analysis." Computers and the Humanities, 10:5. 281-286. Print.
Nardocchio, E. 1986. Computing In the Humanities In Ontario Universities, (report). Print.
Rockwell, G. 2007. "An alternate beginning to humanities computing," (blog), Theoreti.ca. http://www.theoreti.ca/?p=1608.
Schreibmen, S., R. Siemens, and J. Unsworth, eds. 2004. A Companion to Digital Humanities. Oxford: Blackwell. http://www.digitalhumanities.org/companion/.
Smart, J. C., K. A. Feldman, and C. A. Ethington. 2000. Academic Disciplines: Holland’s Theory and The Study of College Students and Faculty. Nashville: Vanderbilt University Press. Print.
Terras, M. 2006. "Disciplined: Using Educational Studies to Analyse ‘Humanities Computing’." Literary and Linguistic Computing. 21:2.