A while back we told you about our acquisition of the record-data cards that were compiled between 1952 and the early 1980s by a prominent group of collectors, dealers, and researchers. One of the most ambitious projects to be spun out of that data is the Grey Gull Matrix Database, which has now been under way for several months.
The first step is to produce something that’s never been seen before — A complete, unadulterated Grey Gull database, using only data that were transcribed verbatim from the original label copy and pressings, with nothing else inferred or added. The card data are remarkably reliable, having been carefully transcribed from the individuals’ own collections as well as a massive research collection that was stored in two New York warehouses. The group were relentless record scavengers, and Grey Gull products were abundant at the time, so there are few gaps aside from some suspected unissued matrix numbers.
Grey Gull has earned a reputation as a discographical quagmire, which isn’t entirely fair. Yes, it was a slipshod operation; but many discographers’ approach to it has been just as sloppy, and some of the perceived problems have been of their own making — the result of their not having taken an objective, systematic approach to the data.
For that reason, data from sources other than the cards and associated record-acquisition sheets won’t be incorporated in the Grey Gull Matrix Database. A random cross-check of existing published data, both printed and online, revealed numerous errors when compared to the cards. The Online Discographical Project, for example, was found to contain significant factual errors (aside from the countless simple typos) in 23% of the entries we cross-checked; and several well-known printed discographies also proved to be less than reliable. Over the years, these works have become so cluttered with guesswork, bad data, and false assumptions — and are so lacking in source citations or other supporting documentation — that it’s become impossible to distinguish fact from fiction. The flawed data often worm their way unchecked into newer publications (as happened in a recently published dance band discography), resulting in what discographer John Bolig characterizes as “fruit from the poisoned tree.” Thus, our decision to bypass all such sources and rely solely on the data cards.
At the moment, data from 800 Grey Gull-related cards have been entered, leaving approximately 3,200 cards and several hundred acquisitions sheets to go. Even at this early stage, the database is yielding valuable insights into Grey Gull’s use of outside masters, its commissioning of outside studios, its relationship with client labels and other record companies, and its operations in general. Yes, it was a messy business, but not to the extent portrayed by some.
The direction the “verbatim” database takes after completion remains open to discussion. To what extent should aural evidence be allowed, for example? To avoid contaminating their primary-source data, the group didn’t enter any aural identifications on the cards, although such experts would have had no problem identifying distinctive singers like Arthur Fields and Vernon Dalhart with 100% certainty. If we do eventually enter aural identifications of anonymous and pseudonymous singers, they will be solicited from collectors and researchers we know to be fully reliable, and will be clearly flagged as being based only upon aural evidence. Some published aural IDs for vocalists, particularly in ODP, are often so far off the mark that we wonder if the claimants ever actually listened to the records.
It’s unlikely that aural identifications will be entered for most pseudonymous and anonymous instrumental sides. The aural evidence is usually far from conclusive — especially with many of the generic-sounding groups Grey Gull employed — and some classic errors have been made over the years in attempting to aurally identify Grey Gull’s bands and their members. Veteran collectors may remember when a British discographer first claimed the cornetist on the 1929 Memphis Jazzer sides was King Oliver, then reversed course with an equally specious claim that it was studio-hack Mike Mossiello, based on Mossiello’s say-so (a claim that’s since been widely questioned, and rightfully so).
Your feedback on the Grey Gull project and its possible direction are welcome. We’ll keep you posted as things progress.