Prewar Blues Lyric Poetry: A Web Concordance


Michael Taft


The Blues Lyric Poetry volumes represent a particularly exciting rite of passage in my life, from a state of ignorance about the blues to a state of understanding. Between 1970 and 1984, I was on a quest to discover the structure that lay behind the genius of blues singers. How did they compose their songs so that each song seemed both new and familiar at the same time? After fourteen years, I had my answer, and I moved on to other questions. But I’m glad that segment of my life produced a few results that did more than simply satisfy my personal quest.


Part of the excitement of that time lay in the relatively new application of computer technology to humanistic studies. I cannot claim to be a pioneer in this area, but the blues concordance that I produced (with the help of Michael Preston, Sam Coleman, and the University of Colorado) may well have been the first extensive computer analysis of popular song lyrics, and certainly the first such analysis of the blues. After key-punching every line from over 2,000 blues lyrics onto individual key-punch cards, and then submitting those cards to the concordance-making software developed by Sam Coleman, I eventually received (I think in 1975) 4,000 large-format green-striped print-out pages of blues concordance. As ugly as it was--and it was ugly, not only in its format, but in the look of the texts as well, with uppercase lettering and all internal codes printed out–it looked beautiful to me.


As I wrote in the preface to the concordance, the “computer concordance is simply a concrete representation” of my intuitions about blues lyric structure, but without this visual–almost tactile–representation, I could never have tested my intuitions. I recall the delightful shock of seeing, for the first time, a line-up of phrases such as “I woke up this morning” and “went to the station” sung by many singers, each singer using the phrase in a different way. In revealing the mechanics of blues composition, the concordance allowed me to analyze that most human of activities–poetry.


In 1977 I completed my doctorate, based on those 4,000 pages of print-out, and as I wrote above, I went on to other questions. I gave the print-out to the Memorial University of Newfoundland Folklore and Language Archive, and stored the key-punch cards in my basement. A few years later, at the urging of Michael Preston, I discussed the idea of publishing the lyrics and concordance with Garland Publications. By that time, Preston was the general editor of a Garland series that published computer-generated literary concordances. Garland agreed to my proposal, and then the work began to resurrect the key-punch cards.


By the early 1980s, Coleman’s concordance software was considerably more elegant than were its earlier incarnations, and Mike Preston’s Center for Computing Research in the Humanities at the University of Colorado was alive with concordance-making. No more klunky uppercase text. No more unsightly codes. We could now sit at a computer console and work directly with the texts stored in the university’s mainframe computers. We could also produce camera-ready copy using a daisywheel printer. By this time I was living in Saskatoon, Saskatchewan, where I edited the key-punch cards. I was one of the last people to operate the University of Saskatchewan’s single remaining key-punch machine.


After completing the key-punching, I loaded my car with over 40,000 key-punch cards and drove for two days to Boulder Colorado, where Mike Preston and I produced the camera-ready copy. By today’s standards, the process was arduous. Editing that gigantic body of blues texts on the mainframe monitor was excruciatingly slow–I would give a command and then make myself some coffee, give another command and then drink a cup. But eventually we were ready to print out the texts which appeared in large font on oversized sheets of white paper (perhaps 17 x 22 inches). These sheets would be reduced to 8 ½ x 11 inches by the printers at Garland.


Proofreading these print-outs was no easy task. Earlier print-outs had allowed me to catch most of the misspellings and typographical errors, but printing with a daisywheel presented other problems: after several hours of use, the plastic in the daisywheel would be so stressed that bits of letters would break off. Proofreading the sheets required that one look for missing serifs from letters or missing dots on “i”s. After more than a week of working almost constantly, the task was done, and Mike Preston packed more than 3,300 pages of large-format print-outs between sheets of plywood, and sent the whole thing off to Garland.


I was quite fortunate to work with Garland, since few publishers would have been prepared to print the entire 3,000 page concordance, plus a 379 page anthology of the lyrics. It is both gratifying and a little painful to realize that the present online concordance could be created in a matter of hours, while the original took years of work. But as I wrote in the original preface to the concordance, it has been in the nature of this project to race along with the technology.


I do wish, however, that I had the time to relisten to the blues transcribed here. Reviewers of the anthology pointed out the many mis-transcriptions in the work, and I have found quite a few myself. Ideally, a work of this sort should never be the responsibility of one pair of ears, but rather the communal effort of those experienced in listening to the blues. I suggested as much at a meeting of blues scholars during the 1975 American Folklore Society annual conference held in New Orleans–that a committee should be formed to create the most accurate transcriptions possible of blues lyrics. That plan would have best represented this great body of American poetry. But the plan never came to pass. I can only suggest, as I did in the original preface to the anthology, that the concordance user listen to the blues songs themselves, as a counterstatement to these transcriptions.




I would like to thank Lars Lindh for taking on the task of converting my 20 year-old computer files into this online blues concordance.