A “doyen” is defined as a senior member of an establishment and, if discography is – or discographers are – an establishment, then Alan Kelly who died in December 2015 was certainly a doyen of discographers.

Alan was born in Girvan on the Clyde in Scotland. He took his MA at Glasgow in 1948 and was later appointed Head of Physics and Head of the Department of Professional Studies at the City of Sheffield College of Education. That College later amalgamated with another to become Sheffield Polytechnic, now Sheffield Hallam University. On amalgamation, Alan and his Department were made redundant and he was thereafter free to pursue other interests.

He had started to collect records in the 1940s and soon developed an interest in opera recordings. As he put it in the Introduction to his Spanish catalogue: “since records were expensive, I also made lists of what was or had been available”. He was still at it more than 50 years later.

He supplied much of the information about red label double sided records issued in HMV’s DA and DB series to J R Bennett who published them under his own name but acknowledged Alan’s contribution in his Introductions. Since then, few lists of HMV recordings have been published without Alan’s keen involvement and those that were, a few even in the august pages of TRC, were generally regarded with disfavour. Long before Larry Lustig took over as editor, in the June/July 1960 issue there appeared a frankly inadequate listing of the recordings of Tito Schipa: matrix numbers and dates were noticeable by their absence but Alan had not been asked to contribute.

40 years ago I sent a note on G&T matrix numbers to James Dennis who published it but only after it was peer reviewed; Alan reviewed it. He wrote me a lovely letter confirming that he, John F Perkins and John Ward had got there before me but let Jim publish it in advance of their definitive article which appeared in March 1974. Alan referred to it as “the archaeology of record collecting as distinct from pot hunting.” Jim Dennis suggested that it would often have to be referred to in years to come: he was not wrong.

Shortly before that, the three musketeers had discovered the “unknown” recordings of John McCormack, unknown because they were issued as recordings by “John O’Reilly”. It could well be those were the records which McCormack used to play to experts explaining that the singer wanted advice on whether he could have a singing career. Only after they had all confirmed that the singer had no chance of any kind of career would he disclose that he had recorded them himself early in his career.

In 1977 they published “Vienna – the first Gramophone recordings” in “Recorded Sound”, the now defunct series issued by BIRS and for many decades “Discography by Alan Kelly” was a by-word for a reliable listing in The Record Collector.

In 1988 he published the first of his numerical catalogues of recordings made between 1898 and 1929 by HMV and its predecessors. It covered Italian recordings and Alan dedicated it to his son, Malcolm. A French volume followed 2 years later and the 1325 page German volume in 1994. Readers of the latter may have noticed the dedication to the physicians and surgeons at the Royal Hallamshire Hospital in Sheffield. The fact is that Alan had developed a brain tumour which they successfully removed. But for what he describes as their “skill and care”, he could not have completed it or the Dutch catalogue. They were the four volumes published in book form by Greenwood Press. Subsequent Russian, Spanish, Czech (including Hungarian), Australian and English catalogues together with listings of matrix number (including unpublished recordings) were issued himself on CD-ROMs. Together with his artists’ discographies, they constitute his discographical legacy.

In 1960 he had married Rita. In his dedication in the English volume, he refers to the fact that she had shared his life for 52 years. Together they cruised the oceans on the QE2 being so well known to the ship’s company that they were given regular upgrades and made to feel part of the ship’s family.

I can, perhaps, describe Alan’s achievement as a response to collectors who wanted to know who had recorded what, where and when. They also wanted to know what unpublished recordings might still lurk in the company’s archive for possible re-issue. Bauer’s 1947 “Historical Records 1898-1908/9” was an important beginning and identified many opera and song recordings made during that period. “Not in Bauer” was thought by dealers to indicate rarity. P G Hurst for many years occupied a similarly God-like status with collectors because of his regular articles in the Gramophone in the 1930’s and 40’s. His “The Golden Age Recorded” was published in 1963 and is still valuable for his personal recollections of the artists he coverd. But he tried to date recordings by the nature of the record labels rather than their matrix numbers. “There is probably a hidden code in the matrix numbers”, he wrote (p.29), “but that is for the future despite much agonising research”.

Alan became a regular visitor both at the British Institute of Recorded Sound and the EMI Archive at Hayes. Leaving his home at Sheffield he would drive down, spend the day going through the Registers and drive home again late in the evening. Regularly. Year after year. Once Kelly, Ward and Perkins had broken the code, Alan’s real work began.

The outcome was not only the breaking of the code but, first, his identification of all issued recordings with artist, title, location and, where possible, the actual date of recording. Because of the numbering system used by the Gramophone Company since its earliest days, Alan was able to compile the catalogues referred to above. His second barrel was then to catalogue each sequence of matrix numbers thereby providing us with chronological lists of everything recorded by the Gramophone Company, not merely in the great European centres but in locations extending to the near and far East. As he wrote in his General Introductions, essentially he came to have four strings to his bow: the Catalogues of all single-sided issue numbers from 1898 to 1929, the Registers for 1929 to 1934, the Matrix Cards identifying who had made each recording and the Coupling Series cards which identified what single-sided or face-numbers had been coupled on each double-sided issue. His achievement was to put it all together in a logical, coherent and digestible form.

With an academic and scientific background he eschewed speculation: it was on the Registers, the weekly returns by the engineers reporting what they recorded and any other available written company documentation that he relied. Nor for him the speculation that if Battistini made some recordings in a BA series that would mean it was a series specifically for him:  Alan’s BA/CA list runs to 477 entries. It was not created for Battistini; it was a series used by Raymond R Sooy, a Victor recording engineer working temporarily in Europe in 1921 which included, by coincidence, Battistini’s May 1921 Milan recordings but which was later used for other recordings in 1929 including over 300 recordings in Teheran and Bagdad. Only if satisfied that other discographers had relied on Victor or Gramophone sources would he include it in his own work. I shall not repeat his blunt description of some other “researchers”. But, in turn, once Kelly established his reputation, the rest of us proclaimed the gospel confident that what he reported was supported by records and the documents in the Archive and that, subject to errors made by office clerks many years before, it could be relied upon. If the documentation was missing or was ambiguous he would say so; but, if he felt able to express a provisional view he would do so making clear it was subject to possible correction if another document or record was later discovered which might provide a different evidence based answer to the particular problem.

In 2007 his work was recognised first by the Association for Recorded Sound Collections in the U.S.A. which conferred on him a Lifetime Achievement Award in a line going back to “Jim” Walsh including Frank Andrews and Bill Moran and, second, by the conferring of an honorary Doctorate of Music by Sheffield University, a first for an English discographer and, possibly, the first world-wide to be so honoured.

He had a great sense of fun and never took himself too seriously. A writer of Bossa Novas, he wrote to me, “was very respectful and hesitant approaching and taking the immensely valuable time of such an authority as I (I like that!)”. The writer had found an error in a Kelly listing which he was the first to accept. He seemed a little worried to be told by his correspondent that his name was frequently taken in vain by a group of web surfers with an interest in 78s. “It seems that the phrase, ‘Kelly says’ … is tantamount to citing Aristotle in a medieval disputation – and I had no idea” he wrote. But the oration when he was given his honorary degree concluded by quoting a recent review, “All serious collectors will welcome Alan Kelly’s outstanding contribution to the professional documentation of a large segment of recorded history, and support this fine work”. I could not have put it better.

JOHN B MILMO

This article first appeared in The Record Collector