Sir Howard Colvin: Architectural historian whose biographical dictionaries laid a foundation for all other scholars in his field

Howard Montagu Colvin, historian: born Sidcup, Kent 15 October 1919; Assistant Lecturer in History, University College London 1946-48; Fellow, St John's College, Oxford 1948-87 (Emeritus), Tutor in History 1948-78, Librarian 1950-84; CBE 1964; Reader in Architectural History, Oxford University 1965-87; President, Society of Architectural Historians of Great Britain 1979-81; CVO 1983; Kt 1995; married 1943 Christina Butler (died 2003; two sons); died Oxford 28 December 2007.

Howard Colvin was the greatest architectural historian of his own time, and perhaps ever. He admired his seniors Sir Nikolaus Pevsner and Sir John Summerson, but both of them were indebted to him for the factual basis on which their judgements were formed; revising Summerson's 1945 Georgian London in 2001, Colvin wrote "[its] combination of brilliant thought and writing with factual carelessness is quite difficult to handle".

The intellectual model whom he regarded as almost faultless was Robert Willis, whose Architectural History of the University of Cambridge (1886) pioneered the solution of archaeological problems by absolute mastery of the documentation, yet Colvin's six-volume History of the King's Works (1963-1982) alone was a greater achievement than Willis's. In addition, Colvin produced what might have remained the authoritative Biographical Dictionary of English Architects 1660-1840 in 1954, had he not expanded it to include Scotland and the years 1600-1660 in 1978 (with the title A Biographical Dictionary of British Architects 1600-1840), and brought out a revised edition in 1995.

It is possible for the very well-informed and very diligent to find an error, or even two, in 1,264 double-column pages of 10-point text, but difficult and unusual. At the time of his death, Howard Colvin had nearly completed proof-reading the fourth version of this astonishing work, whose versions since 1954 have been the starting point of all historical research on the architecture of early modern Britain.

Howard Colvin was the son of Montagu Colvin, a Vickers executive and stamp dealer of lowland Scots descent. Howard won scholarships to more than one public school, but his father chose Trent College, Nottingham, because it seemed to be the best bargain. An urbane and broad-minded history master, Mike Morgan, provided succour in an evidently harsh environment by allowing Howard to visit churches instead of playing games; thus, aged 19, he published his first historical paper, the first of eight on Dale Abbey, Derbyshire, in 1938. He read history at University College London, and in 1943 he married Christina Butler, daughter of the professor of Latin there, sister of the psephologist David Butler, and herself an authority on, inter alia, her Anglo-Irish ancestress Maria Edgeworth.

In 1940 Howard Colvin joined the RAF "because it seemed less bad than the army". Fairly quickly, the RAF lost his records and was unable to do anything with him and a few other unrecordeds except march them from Garstang to Blackpool for lunch and march them back again in the afternoon. On one occasion, lunch was served by a girl whom he had known at university and whose father was an under-secretary in the Air Ministry; she just had time to tell him that her father was looking for archaeologists as photographic interpreters, so in due course he found himself promoted from Aircraftman Second Class to Flight Lieutenant and posted to Malta.

He worked in a limestone tunnel which had an opening in the cliff above the Grand Harbour in Valletta, and during breaks he could use this grandstand to watch the Stukas screaming down onto the British ships at anchor. He recalled the deafening noise of the anti-aircraft barrage and the sight of spent shell cases tumbling from the gun turrets and rolling over the ships' decks until the sailors kicked them overboard.

His unit identified Italian warships from photographs. He was surprised once when a visiting senior officer said "Well done, boy" after one of his identifications, and assumed that this man's evident prior knowledge had come from espionage; only later did he realise that it came from Enigma. He was particularly pleased that he interpreted white lines in dawn photographs as dew on telegraph wires; he persuaded a daring pilot to test this with a low-level photo of Taormina, and convinced his superiors that, as civilian telegraph wires had been dismantled, all the white lines were leading them to Kesselring's headquarters.

In 1946 Colvin was appointed an assistant lecturer at UCL, and in 1948 he obtained a fellowship at St John's College, Oxford, where he remained for the rest of his life, as Tutor (1948-78), Librarian (1950-84), and Emeritus Fellow (1987-2007). As Tutor, he taught the regular Oxford history syllabus, but he managed to add a special paper on English architectural history 1660-1720, then the only form of art history available to Oxford undergraduates, for which he was rewarded by a Readership in 1965.

Oxford respected his productivity and meticulous scholarship, but, for long without art historians of its own, may not have realised that he was even more respected outside its walls; he was never given a chair.

Meanwhile he had a public service career in parallel. He was a commissioner of the Royal Fine Art Commission (1962-72), a commissioner of the Royal Commission on Historical Monuments for England (1963-76), a member of the Historic Buildings Council for England (1970-89), its chairman (1988-89) and chairman of one of its sub-committees (1970-89), a commissioner of the Royal Commission on Ancient and Historical Monuments for Scotland (1977-89), president of the Society of Architectural Historians of Great Britain (1979-81), a commissioner of the Royal Commission on Historical Manuscripts (1981-88), a member of the reviewing committee on the Export of Works of Art (1982-83) and a commissioner of English Heritage (1984-89).

He was an indefatigable attender and a valuable and judicious contributor to discussion until he retired at the age of 70. It was doubtless this unpaid work which was rewarded with his knighthood by John Major's government in 1995; his nomination was supported by two cabinet ministers, the head of the Royal Collection, the chairman of the Royal Commission on Ancient and Historical Monuments for Scotland, a former vice-chancellor, three former and one current head of Oxford houses, two Regius Professors of history, a former director of the V&A and the obiter dicta of the late Sir John Summerson.

Colvin will be most remembered for his scholarly output, unstoppable up to the moment of his death. Although chiefly a historian of architecture, he also wrote two institutional histories (of a religious order and a government department), the history of a profession (architecture), and three histories of localities (Deddington, Holme Lacy and Irford). Unusually he was both a medievalist and a post-medievalist. His first book was The White Canons in England (1951), and the first two volumes of The History of the King's Works (1963) remain the definitive history of the castles, palaces and religious foundations of the medieval kings. He responded to the places to which the war took him with articles on Victorian Malta, Aberystwyth's architecture and Georgian Marlow.

Architecture and the After-Life (1991) covers funerary buildings as far apart as Mesopotamia, Africa, Sweden and Ireland. Although he wrote little about 20th-century architecture, he took a critical interest, and, as a Fellow of St John's, was a patron of Sir Richard MacCormac. He was one of the first architectural historians to appreciate the contribution made by amateur architects, especially in the 18th century, and was one himself, designing an extension to the Senior Common Room at St John's and his own house in north Oxford.

However, the field in which his greatest achievement lies is early modern British architectural history. His three (soon to be four) successive biographical dictionaries both established the methodology and laid the factual foundation on which all other scholars have built. The idea was conceived, he said, in response to the

unscholarly habit of attributing even the most commonplace buildings to one or two well-known architects of the appropriate period. Sir Albert Richardson, then head of the Bartlett School of Architecture, was a prime offender. Seeing himself as an architectural Berenson, he signed certificates of authentication which one used to find hanging in churches and country houses.

Colvin began the first dictionary while still an undergraduate and, surprisingly, was able to work on though the war, finding that the garrison officers' library in Malta had been richly stocked with 18th-century architectural books by the Royal Engineers responsible for the island's fortifications.

Reviewing the work at the launch of the 1995 edition, he claimed that the 1954 edition had no more than 60 stylistic attributions (to several thousand documented ones). "Of these," he wrote:

14 had been confirmed (by documentary evidence) by 1978 and only three proved to be wrong. In the second (1978) edition there were 128 such attributions, of which 14 have since been confirmed and 12 abandoned (though not necessarily shown to be wrong).

Buildings being more abundantly documented than paintings, architectural historians have an advantage over historians of painting, who have to depend more on connoisseurship. But they also have the advantage of Colvin's methodological establishment, his factual groundwork and his example.

In default mode, Howard Colvin's face expressed his formidable powers of concentration, important moments being indicated by balletic eyebrow movements; but this was often replaced by the most engaging smile and occasionally by spasms of abandoned laughter. He and I shared a taste for Georgian grandees with evocative names (Sir Rushout Cullen, Sir Carnaby Haggerston and Hurt Hurt).

He was a tiny man, who loved alpines and had microscopic handwriting. His work was both helped and hindered by innumerable correspondents, either reporting discoveries or seeking endorsements. To keep to his formidable programme, he had no choice but to reply by return, sometimes advising in surprising detail, but always aware of the difference between the significant and the trivial, and particularly alert to "arid art-historical debate". Correspondents reporting what they believed to be a discovery had to be inured to reading "When I saw this drawing in an attic at Blandings Castle in 1950. . ." in reply. His letters are as rich a source of British architectural history as his 132 publications.

Richard Hewlings