forgery and forensics

This page considers forensics, ie determination of the authenticity of objects (including documents and art works) or the information that they embody.

It covers -

• introduction
• principles, context & provenance
• studies
• technologies

It is supplemented by a note on 'security paper'.

     introduction

Determining the authenticity of an object reflects the nature and provenance of that object: essentially what questions can be asked about it and what questions can be answered.

Those questions encompass physical characteristics of objects such as anachronisms in their composition, more subjective matters such as information content or style, and the context in which they are located. The unavailability of substantive information about the history of many antiquities in particular - their lack of provenance - and commercial or curatorial imperatives to ignore uncertainties is a particular problem.

Many forensic technologies will allow determination of an object's age, with varying degrees of accuracy, but necessarily will not tie an object from the requisite period to a particular author.

The preceding pages of this profile have suggested that

• different media, different objects, pose different challenges
• forgery or fraud is often underpinned by a willingness to believe (the complicity of victims and authenticators or other authorities)
• the acceptability of 'unexceptional' items embodies assessments about risk and value
• science is often very good at providing evidence of falsification but poor at conclusively proving authenticity.

For much of history authentication has been a matter of connoisseurship rather than science, with experts relying on intuition (often based on immersion in an author or visual artist's oeuvre), personal idiosyncrasies or minute physical examination to identify discrepancies such as unnaturally regular cracquelure in oil paintings or the tell-tale absence of staining and signs of wear in bound documents.

One rival of art guru Berenson was thus famed for smelling and licking paintings, not unreasonable when the paint on some 'old masters' was barely dry.

More recently a dealer in Cycladic sculptures is quoted as advising

Hold the object between thumb and forefinger and strike it lightly on a doorsill, like a tuning fork. A forgery will emit a clear bell-like ring, whereas a genuine idol emits a dull thump. It is, of course, necessary to experiment with both genuine and imitation idols in order to accustom one's ear to the proper sound

It is only in the past fifty years that authentication has moved out of the library and into the laboratory, although particular figures emphasis intangibles such as 'nose' or 'intuition' (which is presumably restricted an attribute of the individual).

In discussing document forgery James Gilreath commented that

There are some individuals who claim to have what could only be said to be astonishing confidence in their abilities to detect authenticity. In The Hitler Diaries Charles Hamilton boasts that his 'feel test' can distinguish between a genuine and forged document in "two or three seconds" or by inspecting only a "half dozen words". Hamilton on at least one occasion before the publication of The Hitler Diaries described his 'feel test' in his Retail Catalogue A (1984). The first step is to hold the document upside down in order to get a better feel for the writing. As I have written elsewhere, it would be better to hold the document right side up and stand on your head for health purposes in order to gain at least some benefit from this procedure.

Others draw problematical conclusions from data gathered through high tech tools, with claims in 2006 for example that identifying Leonardo da Vinci's left index fingerprint "could help provide information on such matters as the food the artist ate and whether his mother was of Arabic origin".

     principles, context and provenance

As the preceding paragraphs and pages have suggested, forensic examination has two aspects, looking for attributes that successively

i) demonstrate that the item cannot be authentic (eg because it uses an anachronistic technology or features anachronistic information, such as South American fauna in a mediaeval artwork) - exclusionary attributes

ii) associate the item with a particular creator/owner (eg features an authentic signature) - inclusionary attributes

There have been various attempts to reduce forensics to a set of principles. The Bollandist Abbe Mabillon, founder with Lorenzo Valla of the science of diplomatics (ie critical analysis and verification of documents), for example suggested in his 1681 De Re Diplomatica that a document (and by extension other entities) embodied a system of external and internal elements comprising

• acts, ie the determining cause of the document's creation
• individuals who concur in its formation
• procedures through which acts are carried out
• the documentary form that binds the elements together

Notions of original order (respect des fonds) and provenance - differentiating between documents within a functional-structural context and those in isolation (innately less trustworthy, more difficult to authenticate and more likely to be a forgery) - articulated by Mabillon's successors offered both a forensic tool and a mechanism for assessing risk.

That is important since, as we have highlighted in considering questions of electronic authentication and identity, on a day to day basis the need to authenticate persons, personas, items and transactions varies widely. It is not necessary to provide full personal authentication when buying a single carton of milk or using a concession pass on public transport. Other transactions, such as purchase of a $50 million art work, require a greater degree of certainty.

Recognition of a principle of 'best fit' have been reflected in enactments such as Australia's federal Evidence Act and Canada's 1998 Uniform Electronic Evidence Act. They avoid the radical skepticism exemplified by Mabillon's contemporary Jean Hardouin, responsible for the 1697 suggestion that most classical texts were forgeries generated by 13th century monks under the direction of Severus Archontius. It has been echoed by contemporary conspiracist who assert that video footage of the moon landings is a clumsy forgery from a Hollywood back lot.

     studies

As a point of entry for documentation see in Trusting Records: Legal, Historical and Diplomatic Perspectives (Dordrecht: Kluwer Academic 2000) by Heather MacNeil and Scientific Examination of Questioned Documents (Lanham: Rowman & Littlefield 1992) by Ordway Hilton. MacNeil pays tribute to landmark conceptual works such as Muller, Feith & Fruin's 1898 Manual for the Arrangement & Description of Archives.

There is more technical analysis in Kenneth Rendell’s Forging History: the Detection of Fake Letters & Documents (Norman: Uni of Oklahoma Press 1994), Katherine Koppenhaver's Forensic Document Examination: Principles and Practice (Totowa: Humana Press 2007) and Joe Nickell's Detecting Forgery: Forensic Investigation of Documents (Lexington: Uni Press of Kentucky 1996).

Questions about contemporary and historical diplomatics are explored in Luciana Duranti's Diplomatics: New Uses for An Old Science (Lanham: Scarecrow Press 1998), Charles Hamilton's Great Forgers & Famous Fakes: The Manuscript Forgers of America & How They Duped the Experts (New York: Crown 1980) and David Gracy's 'What You Get Is Not What You See: Forgery & the Corruption of Recordkeeping Systems' in Archives & the Public Good: Accountability & Records in Modern Society (Westport: Greenwood 2002) edited by Richard Cox & David Wallace. Duranti collaborated with Terry Eastwood and Heather MacNeil in Preservation of the Integrity of Electronic Records (Dordrecht: Kluwer 2002). Gordon Rugg explores the controversial Voynich manuscript here.

An introduction to art forensics is provided by Otto Kurz's classic Art Forgeries & How To Examine Paintings Scientifically, his Fakes (New York: Dover 1967), Stuart Fleming's Authenticity in Art: The Scientific Detection of Forgery (New York: Crane 1976), Roger Marijnissen's Paintings: genuine, fraud, fake: modern methods of examining paintings (Brussels: Elsevier 1985) and Fakebusters (Chicago: SPIE/McCrone Research Institute 1999) edited by Walter McCrone & Richard Weiss. Joe Nickell's Camera Clues: A Handbook for Photographic Investigation (Lexington: Uni Press of Kentucky 1994) considers photographic forgery and forensics.

     technologies

Forensic technologies essentially aim to exclude rather than include and deal with two aspects -

• exclude the object by identifying anomalies in its physical properties (eg that a 'renaissance' painting includes a pigment not invented until the 1950s or that a 'George Washington' letter is on paper with a watermark first used in 1856)
• provide additional information for an assessment of style or content (eg indicate that the materials used in a work are authentic but that there is an anomaly in the content or other attributes)

Some technologies are non-destructive or non-invasive; others involve damage to the object.

The pigments and inks used by artists and writers broadly reflect the technology available at the time and have thus changed over the centuries (eg from inks based on soot and oak-tree gall to those based on petrochemicals). Laboratory analysis of the pigment can provide an indication of an item's age.

Chemical analysis of paper, canvas or other bearers may also provide indications of a document or graphic work's age. Paper from different eras may for example use grass, linen or timber fibre treated with chemicals or coated with a size that contains whiteners such as titanium and can be detected through chemical analysis in a laboratory.

More prosaically, 'counterfeit-detection pens' used in detection of forged bank notes use an 'ink' that turns gold if the note's paper has the characteristics of genuine paper and turns black if those characteristics are absent.

There is a lucid introduction in The Scientific Examination of Documents: Methods and Techniques (London: Taylor & Francis 1997) by David Ellen.

X-ray (eg x-ray diffraction, infrared microspectroscopy, x-ray photoelectron spectroscopy and x-ray fluorescence) technology can be used to identify the composition of an object or reveal particular features, such as areas of repair or an image that has been over-painted. X-ray analysis has gained popular attention as a tool in the authentication of Old Master paintings, with the technology revealing recent paintings underneath supposed older works, and computer axial tomography offering three-dimensional views of sculptures. It is sometimes accompanied by ultraviolet and infrared examination.

Newer tools such as magnetic resonance imaging (MRI) have increasingly been used in non-destructive viewing inside wooden sculptures, ceramics and bronzes to determine anomalous repair or construction features. Starting points include Paul Mix's Introduction to Nondestructive Testing: A Training Guide (New York: Interscience 1987) and Radiography of Cultural Material (London: Butterworth-Heinemann 2005) by Andrew Middleton & Janet Lang.

Dendrochronology involves dating wooden objects (in particular sculpture and paintings on wooden panels by examination of tree rings. Although dendrochronology registers are now widely available (allowing timber to be dated to a specific year), use is inhibited by requirements that an adequate number of rings be available for examination.

A useful introduction is Tree Rings: Basics & Applications of Dendrochronology (Dordrecht: Kluwer Academic 1989) by Fritz Schweingruber.

Atomic Absorption Spectrophotometry and Inductively-coupled Plasma Spectrometry testing reflects the unique spectral signatures of chemical compositions. Burning of a small sample of a pigment, textile or ceramic glaze can be used to identify its composition and thus determine whether anomalous compounds are present in a work.

Radiocarbon dating employs measurement of the amount of radiocarbon left in an organic object. Radiocarbon, a radioactive form of carbon, is absorbed by animals and plants, decaying at a steady rate once the organism dies. The dating is broadly accurate for objects less than 10,000 years old. Radiocarbon dating of the Vinland Map for example has placed the parchment at around 1434. Testing of the pigment in 1972 suggested that the map was made after 1917; proton-induced X-ray analysis in 1987 indicated that the titanium content of the pigment was far less than in 1972 tests by Walter McCrone and that the map might thus predate 1917.

Stable Isotope Analysis uses examination of isotopes to map stone sculpture to individual quarries, with instances of supposed Cycladic statues being found to come from quarries in the US or Scandinavia.

An introduction is provided by Robert Taylor's Radiocarbon Dating: An Archaeological Perspective (London: Academic Press 1987).

Thermoluminescence, another tool in the examination of ceramics, is based on measurement of light produced when a sample of the object is heated, based on its absorption of cosmic radiation. In principle, the older a work, the greater the thermoluminescence. Test results may be skewed by previous x-ray examination.

Handwriting analysis involves examination of script, ie individual symbols and words. Typically it is based on a close comparison of a questioned document with one of accepted authenticity (eg recognised provenance), with analysts looking for at unique characteristics in the way that letters are formed and the ligatures between symbols.

It has been criticised as a very inexact science - or, unsurprisingly given claims that analysts can provide fortune-teller style insights into the "secrets of life, love and destiny", as a pseudo-science - and received little recognition by courts in most jurisdictions. Defenders comment that some high-profile failures, such as the 1980s Kujau 'Hitler Diaries', were attributable to the lack of valid control documents (ie one forgery was being compared with another, rather than a genuine script).

Points of entry into the literature are Forensic Handwriting Identification: Fundamental Concepts and Principles (New York: Academic Press 2000) by Ron Morris and Questioned Documents: A Lawyer's Handbook (New York: Academic Press 2000) by Jay Levinson.

Textual analysis, pioneered by figures such as Malone and German biblical scholars, uses statistical and other tools in the identification of vocabularies and semantic structures to identify whether a questioned text has the same characteristics as those of one with known authorship. It forms the basis of work about the authorship of some Renaissance literature (eg did Shakespeare write particular plays), in examination of contemporary works such as Primary Colours and - more problematically - in claims about automated determination of the gender of authors of academic papers and email.

Some of the more accessible studies about stylometrics are Computation into Criticism: A Study of Jane Austen's Novels and an Experiment in Method (Oxford: Clarendon Press 1987) by John Burrows, Inference & Disputed Authorship: The Federalist (Reading: Addison-Wesley 1964) by Frederick Mosteller & David Wallace, Analysing for Authorship (Cardiff: Uni of Wales Press 1996) by Jill Farringdon, Vina Tirvengadum's study of the Romain Gary hoax, Joseph Rudman's 2002 overview Non-Traditional Authorship Attribution Studies in 18th Century Literature: Stylistics, Statistics & the Computer and Don Foster's Author Unknown : On The Trail of Anonymous (New York: Holt 2000).

Much handwriting and signature analysis uses stereo microscopes, which provide examiners with a three-dimensional view of ink striations, pen movement and other details. Document examiners often use a video spectral comparator - a digital imaging system that subjects documents to a variety of wavelengths from infrared through ultraviolet - because some inks will become visible and others disappear under different wavelengths, due to differences in chemical formulation.

   next page  (prevention)