Show Summary Details

Page of

 Printed from Grove Art Online. Under the terms of the licence agreement, an individual user may print out a single article for personal use (for details see Privacy Policy).

date: 14 October 2019


  • Molly K. Dorkin

Prior to the 20th century, the attribution of works of art was not governed by rigid regulations, and art dealers and auctioneers assigned attributions based purely on aesthetic grounds. Works were attributed to the artist whose manner they most closely resembled, but they were not further distinguished on the basis of quality; as a result, many paintings purchased as Renaissance masterpieces in the 18th or 19th century have since been downgraded to studio works or even much later pastiches.

Historically, the patrons who commissioned Old Masters placed a premium on subject-matter rather than originality, and popular narratives were requested by multiple patrons, creating conditions in which the demand for copies could flourish (see Copy). Popular compositions were often reproduced many times: by the master himself, an apprentice in his workshop, or even a later follower or imitator. A master trained his apprentices to approximate his manner as closely as possible, and sold the finished work under his own name. In some cases a master would paint the most important part of a work (such as the faces of the central figures) before delegating the rest to apprentices. Through the 19th century, pupils at prestigious institutions were taught by making copies of works by acknowledged masters. Many pieces, particularly drawings (which for much of their history were working tools, rather than art objects), were unsigned. Damaged or incomplete works of art were subjected to extensive restoration or reworking by later artists, a process that can cloud the question of attribution.

For collectors buying from art dealers or at Auction, the name attached to a painting was of greater significance than the correct identification of its author. In the 18th century Grand Tourists filled their country houses with fashionable landscapes by Claude Lorrain, Gaspard Dughet, and Salvator Rosa, often paying as much for skilful 18th-century copies as for 17th-century originals (sometimes knowingly, and sometimes not). Auction catalogues did not distinguish between works by the Old Masters (such as Raphael or Leonardo) and studio productions, stylistically similar works, or copies, and purchasers did not require or expect any proof of Authenticity.

By the early 20th century, auction houses began making some effort to clarify their attributions, although the code was vague: a picture catalogued as ‘Sir Joshua Reynolds’, using the artist’s full name, indicated a more secure attribution than one catalogued simply as ‘Reynolds’. Still, these attributions were made by means of Connoisseurship, a process by which an art historian, dealer, or Auction house specialist would inspect a painting and pass a verdict based exclusively on stylistic grounds. This was an imperfect process as it was completely reliant on the ‘eye’ and honesty of the expert. Many such connoisseurs, wishing to flatter collectors or turn a profit, or willing to be bribed for their opinions, declared paintings of dubious quality to be masterpieces. One such scholar with a controversial reputation was Bernard Berenson, a pioneer of connoisseurship and attribution studies in Renaissance painting. Berenson’s willingness to accept a commission for the works he attributed compromised his credibility, and his published works left a legacy of problematic attributions.

In the modern era, the establishment of a correct attribution is both commercially important and legally required. The attribution of artworks dating from the 20th century onwards is a relatively straightforward matter, thanks to modern record-keeping, which means the Provenance of many modern works can be traced to their origins. Twentieth-century artists did not run large studios or sell works by pupils under their own names. While the authenticity of a work still needs verification, the attribution is more immediately recognizable.

Modern conservators are sensitive in their methods of cleaning and restoring old pictures, not wishing to compromise the artist’s original conception, and attribution research is further supported by a host of scientific means. Conservators use ultraviolet reflectography (UVR) to detect areas of old restoration, x-radiography and infra-red reflectography (IR) to see compositional changes (pentimenti) or underdrawing, and dendrochronology to determine the approximate age of the wooden support of a panel painting. Pigment analysis, which involves both a chemical and visual examination of a microscopic paint sample, can help to establish a terminus post quem for a particular work (see Technical examination).

Scientific methods can be useful in ruling out the possibility that a given work is by a particular artist, but a convincing attribution is almost never possible without the application of connoisseurship and archival research methods as well. The relevant opinion is always that of the author of the most recent catalogue raisonné or, lacking one, a relevant exhibition catalogue or scholarly text. If a work of art is not in a catalogue raisonné, art dealers and auction specialists will consult museum curators or experts currently working on new or revised artist monographs. Few experts will rule on a work of art on the basis of photographs alone; most wish to examine the object in question first hand, as well as to research its provenance. Problems still occasionally arise when a scholar is willing to accept bribes in exchange for a favourable attribution or authentication. Resources such as the Witt photo archives at the Courtauld Institute, as well as the archival records of dealers, museums, and historic collections, are available for consultation. These records can help to establish a work’s provenance, or even the terms of its original commission, more securely. Thanks to advances in scholarship, new discoveries, and scientific innovation, successive generations of experts continue to debate attributions and challenge one another’s opinions.

In the Art market, attribution has a direct bearing on commercial value. Modern auction houses use specific and legally regulated vocabulary, printing a guide to terminology in each sale catalogue. Not only is this approach informative, but it also protects the auction house in the event of dispute or litigation, and the descriptive terms are legally binding subject to a statute of limitations. For paintings, drawings, prints, and works of art, cataloguing in full without qualification indicates that, in the opinion of the auction house specialists, the piece in question is an authentic work by the named artist. A work catalogued as ‘attributed to …’ an artist is considered by the auction house to be probably by the named artist. A work designated ‘studio of …’ is by an unknown hand active in the workshop of the named artist, likely under his direct supervision. (Occasionally, a work will be described as by an artist ‘and studio’, which indicates an unspecified level of workshop participation.) ‘Circle of …’ indicates an artist associated with or influenced by the named artist, albeit indirectly. ‘Manner of …’ or ‘follower of …’ designates a work executed in the style of a given artist, but at a later date. Finally, a work catalogued as ‘after …’ is a direct copy or variant of a known work by the named artist.

For a work to be deemed ‘counterfeit’, as defined by Sotheby’s, it must be an imitation designed to deceive according to authorship, origin, date, age, period, culture, or source in a way that is not reflected by the catalogue description according to the above terms. Moreover, these gradations of attribution are only relevant to works executed prior to the mid-19th century. Impressionist, Modern, and Contemporary works, if not granted a full attribution by the relevant expert or scholarly institute, are declared counterfeit and cannot legally be sold on the open market (see Forgery).


  • E. W. Manwaring: Italian Landscape in Eighteenth Century England (London, 1965)
  • L. Krukowski: ‘A Basis for Attributions in Art’, Journal of Aesthetics and Art Criticism, vol. 39(1) (Autumn 1980), pp. 67–76
  • C. B. Cappell: The Tradition of Pouncing Drawings in the Italian Renaissance Workshop: Innovation and Derivation, 2 vols (PhD diss., New Haven, CT, Yale U., 1988)
  • M. Ainsworth: ‘Implications of Revised Attributions in Netherlandish Painting’, Metropolitan Museum Journal, vol. 27 (1992), pp. 59–76