July 23, 2018





In terms of South African law of evidence, when dealing with the identification of a disputed handwriting, the required evidence need not be given by an expert. However, where the handwriting is disputed and it is attempted to prove the identity of the writer by comparison of specimens, only a handwriting expert can give evidence on the similarities or differences between two specimens of handwriting which are not personally known to them, since the opinion of an unskilled person could add nothing to the court’s own observations, and, of course, it will be the court’s responsibility to make the final decision.

The problems relating to the evidence of handwriting experts is highlighted by Mr Greenfield.  In essence the problems relate to the credibility of such witnesses and the accuracy of their findings.  Mr Greenfield, in his submission, correctly points out that in handwriting, points of similarity do not necessarily prove the identity of the writer and many errors occur because this is often the criteria used by the expert to support his or her findings.  As an example he states that a forged signature by tracing is bound to have many similarities.  In addition a further problem with the evidence is also that the features used to identify similarities fall into different categories, for example, the category of class characteristics, ie features that are common to many people and the category of individual characteristics, ie features which are peculiar to a specific individual.  In addition natural variations are bound to occur in everybody’s handwriting, and, for an expert bent on finding dissimilarities to prove that the features are in fact dissimilarities, is yet another source of error.

In his view the greatest source of error arise from the inadequate number of known specimens (standards) available for comparison purposes.  Here again standards fall into categories, namely collected standards which are writings that appear on documents written in the normal course of events and standards which are of lesser value, ie writings which are specifically requested for comparison purposes.  In addition Mr Greenfield is of the view that one of the sources of error is the failure to investigate fully the line quality (which includes the smoothness of the line and the pressure pattern), the pen movement, the line direction and the sequence of strokes.  At present there is no governing body that determines who is an acceptable handwriting expert or witness in a court of law and as a result anybody who feels he or she knows something about handwriting analysis can present him or herself as a handwriting expert.  In addition he states that there are several primadonnas about, who have spent many years in the field, and who have fixed approaches regarding methods that are not always correct and they are not prepared to yield.  The result is that the profession concerned (a forensic examiner of questioned documents) is degraded and treated with contempt in our courts of law.


Mr Greenfield proposes a solution to the problems identified above and, in his view, all suitably qualified “Forensic Questioned Document Examiners” who practice the profession on a full-time basis, should be invited to attend an inaugural meeting, convened by the Department of Justice in order to present their credentials confirming their expertise in the field.  A register of all acknowledged practitioners should be drawn up and administered by the Department or a committee appointed by the Department.  Practitioners should be required to register with this body and to pay a registration fee.  A code of Ethics should be developed to guide their conduct and an ombudsman, comprising three registered practitioners, drawn from the registered members on a roster basis, should be appointed to deal with situations where there are conflicting opinions, and the consensus opinion of the ombudsman should be accepted by both parties.  The expert accepted by the ombudsman will be the one to give evidence in court and he or she present the consensus opinion.






The evidence of experts in South African law of evidence presents peculiar difficulties in the assessment of its probative value.  “Expert witnesses”, remarked Ramsbottom J in R v Jacobs[1]  “are not the judges of fact in relation to which they express an opinion” and judicial officers “should be careful, therefore, not to allow the opinion of witnesses to take the place of their own finding of fact”.  The court does not usually have any means by which it can verify the witness’s conclusions, and if there is a conflict of expert testimony in some matters where the rationale for the opinion is utterly beyond the grasp of the usual trier of fact, it may be thrown back upon doubtful factors such as the rival witnesses’ reputation and experience.  A court which relies on the evidence of an expert is, to a greater or lesser extent, at times taking a step in the dark – something which should be done only with considerable caution.  Usually the determination depends on the examination of opinions and the analyses of the reasoning behind them.  A good deal will depend upon the general repute of the witness’s profession.  It is generally accepted that the identity of fingerprints can be accurately established by expert inspection even though the points of identity may not be apparent to an untrained person, and a court may therefore decide that it is safe to accept an expert’s opinion on the matter despite being unable to satisfy itself that the prints are identical.  This, however, is not the case with the evidence of handwriting experts in South Africa, as will be argued below.


In terms of South African law any witness who knows a person’s handwriting is competent to identify it.[2]  He or she may have acquired this knowledge by observing the person writing or by receiving documents which purported to be in his or her handwriting, or seeing such documents in the ordinary course of business.  The degree to which the witness is familiar with the disputed handwriting will affect the weight of his or her opinion, but even the fact that he or she has seen the writing only once does not render his or her evidence inadmissible.

More difficult problems arise when attempts are made to identify a handwriting by comparison with another specimen which has been proved to be genuine.  Our courts have frequently emphasised that this method of identification must be used only with the greatest caution.  A witness who is looking for similarities in two specimens of handwriting is unlikely not to find any, and this may involve the witness in an unconscious circuitry reasoning.

There are, however, statutory provisions which expressly allow the courts to rely upon a comparison of handwritings.  Section 228 of the Criminal Procedure Act, 51 of 1977, reads as follows:

Comparison at criminal proceedings of a disputed writing with any writing proved to be genuine, may be made by a witness, and such writings and the evidence of any witness with respect thereto, may be submitted as proof of the genuineness or otherwise of the writing in dispute.[3]

Only a handwriting expert is entitled to give evidence on the similarities or differences between two specimens of handwriting which are not personally known to him or her, since the opinion of an unskilled person could add nothing to the court’s own observations, and, of course, it will be the court’s responsibility to make the final decision.[4]  But the statute has been interpreted to mean that the court may rely upon its own comparison of the writings, unassisted by expert evidence, although, as a matter of practise such a course is greatly discouraged.  In R v Kruger[5], after what was called by the Supreme Court of Appeal  in S v Boesak an exhaustive review of the comparative English position, De Beer I said:

S 248 [which was to the same effect as the present s 228] to my mind empowers the court to compare documents without the intervention of expert witnesses or handwriting students.  But in doing so it is almost impossible to overstate the attendant dangers..

In S v Boesak[6]  the Supreme Court of Appeal approved the rule as stated in R v Kruger (supra) as being correct in principle and remarked that the position in our law is, in essential respects, similar to that in the United States, Canada, and Australia, as well as England. In particular the Court stated the fact that the court itself is allowed to compare the handwriting of the appellant on the letter with other genuine specimens of his signature, is acknowledged in our law, as in several other legal systems. This was laid down by the Full Bench of the Orange Free State in R v Kruger 1941 OPD 33 at 38, after an exhaustive review of the comparable position in England. (See also s 228 of the Criminal Procedure Act 51 of 1977.).  The Supreme Court of Appeal stated that the rule seems to be correct in principle. Even in cases where expert witnesses testify, it is the Judge who bears the responsibility of making a final judgment (Coopers (South Africa) (Pty) Ltd v Deutsche Gesellschaft für Schädlingsbekämpfung MBH 1976 (3) SA 352 (A) at 370E – H; Gentiruco AG v Firestone SA (Pty) Ltd 1972 (1) SA 589 (A) at 616D-617C. See also Hoffmann and Zeffertt The South African Law of Evidence 4th at ed 104-6.) The position in our law is, according to the court, in essential respects, similar to that in the United States (Wigmore On Evidence paras 21-29 et seq); Australia (Adami v The Queen (1959) 108 CLR 605 (High Court of Australia)  D  at 616-7); Canada (R v Abdi (1997) 34 OR (3d) 499 (CA)) and England ( R v Rickard (1918) 13 Cr App R 140; Cross and Tapper on Evidence 8th ed at 761, and Phipson Evidence 14th ed paras 17-15 and 17-16.) The court stated that the rule under discussion should be applied with caution, but, taken in conjunction with all the other factors indicative of the authenticity of the letter discussed above, the court was of the view that it was entitled to conclude, prima facie, that it was written and signed by the appellant. In the absence of evidence to the contrary, and having regard to all the other indicia mentioned above, the court was satisfied that the authenticity of the letter has been proved beyond reasonable doubt.

In the appeal to the Constitutional Court (S v Boesak 2001(1) SA 912 CC) the Constitutional Court found no fault with the fact that the Supreme Court of Appeal had made such comparison on appeal but after the hearing of the appeal.[7]  This is so because the comparison is merely part of the evaluation of the totality of the evidence. Taken in conjunction with all the other factors indicative of the authenticity of the letter, the Constitutional Court was of the view that the Supreme Court of Appeal was entitled to conclude prima facie that the letter was written and signed by the appellant. Together with the other circumstances the Constitutional Court was satisfied that the State had proved beyond reasonable doubt that the whole amount donated by Simon had been donated to the children of South Africa and that the appellant had unlawfully appropriated R250 000 of that amount. In the result the appeal against the convictions on counts 4 and 5 were dismissed.

In S v Van Dyk [8] the court held that the evidence of a handwriting expert is to be approached with caution. A handwriting expert testified that seven points of correspondence between the undisputed specimen handwriting and the disputed handwriting were sufficient to identify the latter as having been written by author of former. This evidence was not challenged.  The accused was convicted where seven points of correspondence were established. Furthermore, a handwriting expert compared the forged signature with undisputed examples of the appellant’s handwriting, and concluded that the signature had been made by the appellant. The expert indicated seven points of correspondence on a chart, and mentioned in his evidence that there was, in addition, an eighth such point. According to the expert, seven points of correspondence were sufficient to identify the appellant’s handwriting beyond reasonable doubt.





The problems highlighted by Mr Greenfield relating to the evidence of handwriting experts are not peculiar to South African law.  Similar problems and concerns exist for example in the USA. Michael J. Saks[9] gives the following background to the evidence of handwriting experts in the USA:


The opinions of experts upon handwriting, who testify from comparison only, are regarded by the courts as of uncertain value, because in so many cases where such evidence is received witnesses of equal honesty, intelligence and experience reach conclusions not only diametrically opposite, but always in favor of the party who called them.

While some courts continued to reject such expertise, and most that allowed it remained sceptical, a group of professional experts was growing up and beginning to seek greater respectability. It is ironic that when expert handwriting identification testimony was first declared admissible in America and England, there were no experts. That is to say, the lawyers seeking to admit such testimony merely had to proffer various witnesses who were willing to assert a kind of ad hoc expertise acquired as a side effect of being something else, such as a postal inspector or a bank teller. No practicing forensic document examiner today would concede any expertise to such witnesses.

When the legal system agreed to accept handwriting identification testimony, however, it created a demand which was met by people who increasingly turned their entire attention to filling it. Not surprisingly, these people soon set out to create a standard theory and practice, giving their trade the appearance of “science.” Among the first of those people was Charles Chabot, who, despite his name, was English. Originally a lithographer by trade, he developed an interest in handwriting identification about the time such expert testimony was gaining admissibility in English courts. It is unclear how much he was influenced by contemporary French theory and practice, but in 1871, at the urging of his lawyer-disciple Edward Twistleton (who wrote a lengthy theoretical introduction to the book), Chabot published The Handwriting of Junius Professionally Investigated. This was the first book in English to assert that there was a science of handwriting identification, and to illustrate its methodology.

Two American books on handwriting identification were published in the 1890s–William E. Hagan’s Disputed Handwriting (1894) and Daniel T. Ames’s Ames on Forgery (1899). But the event that was to transform handwriting identification expertise from ugly duckling to swan was the 1910 publication of Albert S. Osborn’s Questioned Documents, with an introduction by John Henry Wigmore.

Osborn’s book, Osborn’s personality, and Osborn’s friendship with Wigmore were the cornerstones upon which respect was built for handwriting identification expertise in the United States. ] Osborn set out the theory and practice of the claimed expertise so comprehensively that it is fair to say that all treatments of the subject since have simply been rearrangements or expansions of Osborn’s 1910 book. As to his personality, he was clearly a man of exceptional intelligence and critical abilities, but with a blind spot. He had a kind of mystical faith in the ability of the human mind to create a system of analytical expertise for the solution of virtually any class of problem. While he could be laudably sceptical regarding the claims of others, he never seemed to notice that most of the generalities upon which he built his system lacked empirical verification. Nevertheless, he had faith in his vision and his ability to sell others on that vision, whether the audience was a jury or a group of students, lawyers, or judges. His most significant convert was Wigmore, the most influential figure in evidence theory in the last century. Together, Osborn and Wigmore conducted a quarter-century public relations campaign on behalf of “scientific” handwriting identification expertise as practiced by Osborn and described in his book.

The ultimate triumph of this vision was finally insured by the Lindbergh baby kidnapping case, State v. Hauptmann, in 1935. Osborn was the chief witness called to testify that Bruno Richard Hauptmann had written all of the ransom notes found or sent after the abduction of the son of Charles A. Lindbergh. The public seemed to need to believe Hauptmann was guilty, wanted him convicted, and was grateful to those who supplied the evidence. Osborn became a celebrity. For nearly sixty years after the affirmance of State v. Hauptmann, no reported opinion rejected handwriting expertise, nor displayed much skepticism towards it. The testimony, which at the turn of the century was deemed of “uncertain value,” became universally regarded as scientific and dependable. In 1977, a New York court noted the change: “Since that rather cynical observation was made by our highest court in Hoag, examiners of questioned documents, as handwriting experts prefer to be called, have attained more respectable standing in the courtroom.”  As a New Jersey court observed in 1957, after the Hauptmann case, handwriting identification expertise could no longer be regarded as “the lowest order of evidence, and … accorded little evidential weight.”

Frye and Daubert

For generations judicial thinking concerning the required dependability of expert testimony, especially that which might be labelled “scientific” testimony, was dominated by the so-called Frye test. The essence of the test was that testimony concerning scientific expertise was admissible only if the validity of the scientific principle or process upon which it was based had obtained general acceptance in the relevant scientific community. Unfortunately, Frye itself contained no guidance on how to determine what constituted the relevant community to be looked to for acceptance. (This is hardly surprising, considering that the Frye opinion itself was less than a page and a half long.) In addition, Frye suggested that this test was required only for “novel” scientific evidence, without explaining why the same test was not appropriate for older claims and methodologies.

For nearly seventy years, judicial and academic exegetes made virtually whatever they wished out of the Frye test. Some found it to be a formidable barrier to admissibility, and others the most illusory of restrictions on the introduction of unvalidated and undependable “expertise.” Critics of the Frye test attacked it from both sides, some saying its approach should be abandoned because it kept too much out,  and yet others saying that it should be abandoned because it let too much in.

When Congress adopted the Federal Rules of Evidence in 1975, it made no reference to Frye, either in the language of Federal Rule of Evidence 702 (FRE 702) itself or in the short and particularly unhelpful advisory committee note. In the almost two decades that followed, courts and commentators variously construed (FRE 702), the standard it implied, and the judge’s role in enforcing it, including whether any of the many versions of “Frye” continued to play any proper role at all in federal court trials under the Federal Rules of Evidence. Initially, proponents of the “let it all in” school of thought seemed to dominate under a broad construction of what might “assist the trier of fact to understand the evidence.” But by the early 1990s, there was increasing sentiment that judges were admitting too much expert testimony of little or no dependability. In 1991, this led the Judicial Conference Advisory Committee on Civil Rules to propose an amendment to (FRE 702) which would have required that all expert testimony–not just “scientific” expertise–be subject to a preliminary finding by the judge that it was “reasonably reliable.” While this rule was never promulgated, it formed the immediate background for the Supreme Court’s decision in Daubert v. Merrell Dow Pharmaceuticals, Inc.

In United States v. Lewis,[10] the primary question posed by the defendant’s motion to prohibit the testimony of forensic document analyst John W. Cawley, was whether his handwriting identification evidence was sufficiently reliable to be admissible pursuant to Rule 702 and Daubert.[11]  The Government argued that the court was not required to apply Daubert to handwriting identification analysis and that, in any event, Mr. Cawley’s testimony was reliable.

The court noted that Rule 702 provides (see footnote 10):

If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise, if (1) the testimony is based upon sufficient facts or data, (2) the testimony is the product of reliable principles and methods, and (3) the witness has applied the principles and methods reliably to the facts of the case.

As the Supreme Court explained in Daubert and Kumho Tire, under Rule 702, the district judge must ensure that the expert’s testimony is both relevant and reliable before it may be admitted, regardless of whether the testimony is scientific or based on technical or other specialized knowledge.[12]. When the expert’s testimony’s “factual basis, data, principles, methods, or their application are called sufficiently into question, the trial judge must determine whether the testimony has a reliable basis in the knowledge and experience of the relevant discipline.”[13]

The court concluded that:

In performing this gate-keeping responsibility, the Supreme Court has articulated four factors the court may consider:

(1) Whether a theory or technique can be or has been tested;

(2) Whether it has been subjected to peer review and publication;

(3) Whether, in respect to a particular technique, there is a high known or potential rate of error and whether there are standards controlling the technique’s operation; and

(4) Whether the theory or technique enjoys general acceptance within a relevant scientific community.

526 U.S. at 149-50(citing Daubert, 509 U.S. at 592-94) (internal quotations marks and alterations omitted). These various factors are not an exhaustive list of all possible ways to assess reliability, nor must all of the factors be applied in every case. 526 U.S. at 150. Depending on the facts of the case and the type of testimony being challenged, it may very well be unreasonable to apply all of these factors. Id. at 151. Accordingly, the trial judge is given discretion in determining how and in what manner to make reliability determinations pursuant to Daubert.

Where, however, the Daubert factors are reasonable measures of the testimony’s reliability, the Supreme Court has instructed that the trial judge should consider them. Id. at 152. While district courts have considerable leeway in determining how to assess reliability, they do not have the discretion to simply abandon their gate-keeping function by foregoing a reliability analysis. Id. at 158-59. (Scalia, J., concurring). Significantly, “in a particular case the failure to apply one or another of [the Daubert factors] may be unreasonable, and hence an abuse of discretion.”   Id. (Scalia, J., concurring).

This court is not persuaded by the Government’s argument that the court need not apply the Daubert factors. For support, the Government cites several cases where circuit panels have affirmed a district  court’s admission of a handwriting expert. See, e.g., United States v. Jolivet, 224 F.3d 902, 905-06 (8th Cir. 2000) (affirming under plain error review the district judge’s decision to admit handwriting expert’s testimony without applying Daubert factors); United States v. Paul, 175 F.3d 906, 910-11 (11th Cir. 1999) (concluding summarily that Daubert factors do not necessarily apply to admission of handwriting expert, but failing to address any alternative measures of reliability); United States v. Velasquez, 33 V.I. 265, 64 F.3d 844, 849-50 (3d Cir. 1995) (cautioning against strict application of the reliability requirement and finding that the ultimate touchstone is helpfulness to the trier of fact).

While these cases all emphasize the district judge’s discretion in choosing how to assess the expert’s reliability and the “flexibility” afforded the court under Daubert and Kumho,   they do not provide any rationale for declining to apply the Daubert factors. These courts simply downplay their rejection of the Daubert factors by minimizing the importance of reliability. The rationale given in these cases is that reliability in the handwriting identification context is a less significant concern due to the fact that the jury is able to see for itself whether the writings are similar. See, e.g., Paul, 175 F.3d at 911; United States v. Jones, 107 F.3d 1147, 1160-61 (6th Cir. 1997) (citing United State v. Buck, 1987 U.S. Dist. LEXIS 9913, No. 84 Cr. 220-CSH, 1987 WL 19300, at *3 (S.D.N.Y. Oct. 28, 1987)).

The Supreme Court’s mandate in Daubert, however, runs contrary to this rationale. There, the Court explained that Rule 702’s requirement that evidence “assist the trier of fact in reaching its conclusion” goes primarily to relevance; an assessment of reliability is an additional component of the judge’s gatekeeper function. Daubert, 509 U.S. at 591-92. Simply put, expert testimony that does not relate to any issue in the case is not relevant, and thus, not helpful. Reliability, on the other hand, is an assessment of whether the expert’s reasoning or methodology is valid and warrants the relaxation of the common law first-hand knowledge requirement for witnesses. See id. at 592.

Here, the court finds that all of the Daubert factors reasonably apply to handwriting analysis and thus are helpful to the court in assessing the reliability of Mr. Cawley’s testimony. As a branch of forensic science, handwriting analysis has many characteristics that are resonant with the traditional concept of “science.” See Reference Manual on Scientific Evidence, Federal Judicial Center 69 (2000) (noting that “science” embodies the vast array of knowledge about the mysteries of our world, including the technologies which have transformed our lives). Handwriting analysis proposes a theory that each person’s handwriting is unique, and involves a method by which a trained expert can identify each writing’s author. The sufficiency of that theory and method can be tested through the basic factors set forth in Daubert

For instance, because the results in handwriting analysis are based on identification, there must be a corresponding probability of error. See D. Michael Risinger & Michael  J. Saks, Science and Nonscience in the Courts: Daubert meets Handwriting Identification Expertise, 82 Iowa L. Rev. 21, 36 (1996). In other words, it is possible to calculate the number of times a handwriting expert correctly identifies the author of a handwriting sample. This number can then be used by courts as an indicative error rate. Other qualities of handwriting analysis, such as the theory that penmanship characteristics are separable from each other, and that there is a base rate of penmanship characteristics in a population of potential authors, are also capable of measurement. Id. at 36-37.

Yet despite the relative ease with which such measurements could be made, the Government did not offer any evidence of reliable testing and error rates, or of any of the other Daubert factors through Mr. Cawley’s testimony. The Government had the burden of establishing by a preponderance of the evidence that Mr. Cawley’s expert testimony was sufficiently reliable to be admissible under Rule 702. See Maryland Cas. Co. v. Therm-O-Disc, Inc., 137 F.3d 780, 783 (4th Cir. 1998). The court found that the Government did not meet  its burden.


In a recent article Michael J Saks[14] pointed out that document examiners do many things. They examine typewriting for signs of idiosyncratic typeface wear and alignment that might indicate a common origin for two documents. They analyze ink to reveal its physical and chemical properties. They scrutinize the alignment of printed lines and the overlap of handwritten lines to determine if words or phrases have been after-inserted. And they analyze the composition, method of production, and watermark of paper to ascertain its probable origin and, in some cases, its age. In performing these tasks, the document examiner may often use specialized knowledge of manufacturing processes and manufacturer specifications, not unlike that employed in firearms identification concerning the relative number, spacing, pitch, and direction of twist of grooves and lands in various makes of rifled barrels.

He highlights the fact that two events have occurred in recent years which combined to stimulate a re-evaluation of handwriting identification expertise. The first was the 1989 publication of an article in the University of Pennsylvania Law Review, pointing out the lack of empirical validation of the claims of the expertise. The other was the U.S. Supreme Court’s 1993 decision in Daubert v. Merrell Dow Pharmaceuticals,[15] Inc., which rejected previous approaches to the acceptability of scientific expertise under the Federal Rules of Evidence, and put into play the validity of claims of scientific expertise even in areas in which that expertise had long been taken for granted. As a result there has been a flurry of litigation over the validity of handwriting identification expertise, resulting in several recent federal decisions.

In another article Bob E. Lype highlights the problems with the evidence of handwriting experts in the following terms:[16]

Handwriting expert testimony has been admitted in courts throughout the United States for well over a century and a half, and in Tennessee handwriting experts were specifically permitted to testify by statute from 1889 until 1991. When the “Frye test” was the accepted standard for admitting expert testimony, handwriting experts easily met the “general acceptance” standard. However, a number of recent events have raised new questions about the role of handwriting experts and the admissibility of their testimony, both nationwide and in Tennessee. Whereas it was once generally accepted that a lawyer faced with a questionable signature or handwritten document could simply “get a handwriting expert” to resolve the issue, today if a lawyer’s case turns upon the genuineness or authenticity of handwriting, that lawyer must do substantially more homework regarding handwriting experts. If a handwriting expert becomes involved in the case, the lawyer should be prepared to defend (or attack, depending upon the perspective) handwriting expertise.

The Recent Developments

The profession of handwriting experts, sometimes also known as forensic document examiners or questioned document examiners, has been under a full-scale attack nationwide in recent years, and the attack is making its way into Tennessee courts. The attack began with a scathing law review article published in 1989 in which the authors concluded that (1) there was insufficient testing of the proficiency and methodology of handwriting experts to establish the reliability of their expert testimony, and (2) the research and data that existed at the time tended to show that handwriting experts were not nearly as accurate or proficient as they claimed to be. Shortly thereafter, the Supreme Court overruled the general acceptance standard of the “Frye test” when the Daubert decision appeared in 1993. The Daubert decision is sometimes viewed as the Supreme Court’s answer to the proliferation of “junk science” testimony in the courts, and it imposed upon the federal district courts a gatekeeper role in assessing the reliability and methodology of scientific expert testimony. The federal courts are split as to whether Daubert applies to nonscientific expert testimony (i.e., testimony based upon “technical or other specialized knowledge,” rather than science, as contemplated by Rule 702). While the Supreme Court clarified the Daubert standard in late 1997, it did not address the application of Daubert criteria to non-scientific expert testimony.

Following the Daubert decision, some of the authors of the 1989 law review article were called to testify as expert “critics” of the reliability of handwriting expertise, in an attempt to disqualify professional handwriting experts. While this tactic met with only limited success, it became increasingly popular. Perhaps the pinnacle for the “expert critics” of handwriting examiners occurred in 1995, when the Third Circuit reversed a criminal conviction because the district court failed to permit one of the law review authors to testify as an “expert critic” of handwriting analysis. Since then, the “expert critics” have been offered in numerous cases around the country, and there is a growing body of case law on the federal district court level addressing their contentions and the admissibility of handwriting expert testimony.

The Tennessee Supreme Court gave us the McDaniel decision in late 1997, declining to specifically adopt the Daubert standard for the admissibility of expert testimony under the Tennessee Rules of Evidence, but nonetheless imposing a similar gatekeeping role upon Tennessee trial judges in assessing the underlying reliability of scientific expert testimony. As in the federal courts, there remains a question as to whether the McDaniel standard applies to non-scientific testimony based upon “technical or other specialized knowledge,” but the Tennessee Supreme Court did note that in deciding whether expert testimony will “substantially assist the trier of fact” for purposes of Tenn. R. Evid. Rule 702, the trial court must necessarily make a determination as to the validity or reliability of the evidence. This statement would arguably be true regardless of whether the evidence is scientific or based upon “technical or other specialized knowledge.”

Is Document Examination a Science?

Document examiners frequently refer to themselves as “forensic scientists.” On a larger scale, some would question whether any of the “forensic sciences,” such as fingerprint identification, firearms or bite mark identification, or even forensic pathology, are “true sciences.” As the courts decide whether Daubert and McDaniel apply to non-scientific expert testimony, this may become a more important concern. Can handwriting expertise be held to scientific standards?

From the lawyer’s perspective, it makes little practical difference whether a handwriting expert is permitted to testify under the scientific or the “technical or other specialized knowledge” arm of Rule 702. This may affect how the lawyer is permitted to refer to the expert’s testimony in argument, or whether the expert may qualify his opinion “within a reasonable degree of scientific certainty.” It may also lead to a special jury instruction from the court. While this distinction involves more than mere semantics, and it is no doubt important to the handwriting experts, the lawyer’s first concern is whether the expert will be permitted to testify before the jury.

While handwriting experts have offered mixed answers to the question of whether their expertise is “scientific,” there is a trend in the courts to find that they are qualified because of “technical or other specialized knowledge.” In U.S. v. Starzecpyzel, the district court in New York gave an extended discussion of this question. The court heard testimony from one of the co-authors of the 1989 law review article criticizing the reliability of the anticipated expert testimony, but it permitted the handwriting expert to testify after analogizing the expert’s work to the non-scientific work of a harbor pilot who has repeatedly navigated a particular waterway. This same conclusion was reached more recently by the Sixth Circuit in U.S. v. Jones. In that case, the court noted that no other courts have found expert handwriting analysis inadmissible under the Federal Rules of Evidence. In deciding that the handwriting expert was qualified, the court noted that scientific principles relate to aspects of handwriting analysis, but the expertise is not truly scientific.

The question of whether handwriting expertise is scientific may become more or less significant as the courts decide whether the Daubert and McDaniel criteria apply to non-scientific expert evidence. The present trend seems to favor admitting the testimony as non-scientific, but the “expert critics” are continuing their attacks on the field.

Reliability Concerns

The primary thrust of the attack on handwriting expertise concerns reliability. The critics claim there has not been sufficient testing of the capabilities of document examiners, and that the data available shows they have a surprisingly poor accuracy rate. It is generally conceded that the 1989 law review article correctly pointed out the dearth of reliable test data in the field, but there is a great disagreement over the interpretation of the data that exists.

While there have been sporadic tests and studies of document examiner proficiency through the years, the central debate is over the meaning of two groups of tests — five tests mailed out by the Forensic Sciences Foundation (FSF) in the 1970s and ’80s, and a group of related, controlled tests performed by Dr. Moshe Kam in the 1990s. The critics of handwriting expertise claim that the FSF tests demonstrate extremely poor accuracy rates by document examiners, concluding (and sometimes testifying) that they establish at best a 57 percent correct, 43 percent incorrect “batting average” for the group. On the other hand, proponents of handwriting expertise are quick to point out serious deficiencies in the methodology, participation rates and reliability of the FSF tests, which lacked control groups and were mailed out to anyone who ordered them. Proponents of handwriting expertise point to three recent studies by Dr. Kam, which they claim demonstrate that handwriting experts possess skills far superior to laypersons. Dr. Kam’s controlled tests compared the abilities of document examiners and laypersons in correctly matching various sets of handwriting. Critics of Dr. Kam’s studies challenge his interpretation of the test data and the make-up of his tests. They claim that the non-professional test-takers were not motivated to perform, that the tests were too easy, and that the proper interpretation of the test data is that laypersons did just as well as professionals.

The critics of handwriting expertise also complain that the field lacks standards, discipline, and a true methodology. They argue that handwriting experts do not quantify their findings with any numerical analysis of similarities or differences in compared writings, and there is no uniformity in procedure, nor even terminology. The critics also complain that the recognized bodies certifying document examiners are akin to a fraternity, and there is no meaningful testing in the certification process.

Proponents of handwriting expertise, on the other hand, claim that they do follow a recognized methodology and protocol. They also note that there is a recognized certification board for document examiners, and that board-certified examiners have completed an extensive two-year training program and a vigorous certification process, complete with written and oral board examinations. Finally, the handwriting experts point out that not all fields of expertise are subject to numerical standards. Unlike fingerprint analysis, there is no “magic number” of similarities or differences which will permit a handwriting identification. The handwriting experts note that, while fingerprints are static and essentially unchanging, free and natural handwriting will change according to a variety of influences.

These disagreements have led to several recent cases where Dr. Kam and one of the co-authors of the 1989 law review article have given extensive expert testimony, either in connection with a motion in limine, a Daubert hearing, or (recently in Tennessee) a McDaniel hearing. These hearings are typically filled with complicated and confusing statistical interpretations, as well as attacks on the credibility of the opposing experts, so they are “battles of the experts” in the truest sense. Until there are binding decisions from higher courts, it appears that these battles will continue, as the critics of handwriting expertise show no signs of lessening their attacks.

The Role of “Expert Critics”

One aspect of the current debate that is sometimes overlooked is the proper role of an “expert critic” of a particular field. In order to give opinions regarding the reliability and deficiencies of handwriting expertise, or any other field, the critic must first be qualified as an expert himself. How does one qualify as an “expert critic?” This remains unclear and will no doubt lead to further litigation. With regard to the co-authors of the 1989 law review article, none of them has been trained or tested in the principles of forensic document analysis. They are academic professors, and their knowledge of the field is through self-directed self-study, including research of the legal history of handwriting experts and the published studies of proficiency tests, together with anecdotal discussions with handwriting experts. Two of the co-authors have no training in statistical analysis or research methods, yet they have been permitted to testify regarding interpretations of the reliability and meaning of various proficiency tests.

While it is clear that there must be some inquiry into the reliability of expert testimony under Daubert and McDaniel, the question remains how one becomes qualified as an “expert critic” of any field. While this is presently at issue in the arena of handwriting expertise, it will likely become an issue in other fields as well. Some of the “expert critics” have hinted that they will soon turn their scrutiny upon other forensic sciences.


The testimony of handwriting experts has been generally accepted for decades. As a result, their testimony has been highly influential in criminal convictions and acquittals, as well as civil cases upholding or rejecting wills and deeds. The courts that have considered the recent attacks on the admissibility of handwriting expert testimony seem to favor admission of the testimony as “technical or other specialized knowledge.” These courts are no doubt influenced by the long history of general acceptance of handwriting expert testimony as reliable, which the document examiners claim is bolstered by the recent test data compiled by Dr. Kam.

The current battles spring from noble intentions, asking the courts to scrutinize the abilities of handwriting experts before they are given the weighty mantle of “expert witness” before a jury. Now that the cards are being placed on the table, there is a tremendous controversy over how to interpret all of the data. Until higher courts ultimately resolve this question, lawyers in Tennessee and elsewhere must take a closer look at the role of handwriting experts, and if necessary, must join the battle.22


For a long time forensic handwriting analysis seemed more respectable, but its status has been shaky since 1993, when the Supreme Court handed down its ruling in Daubert v. Merrell Dow Pharmaceuticals. Previously the chief criterion for the admissibility of expert testimony had been whether it was based on techniques “generally accepted” by scientists. Daubert gave federal judges much greater discretion in deciding admissibility. It suggested they consider (1) whether a theory or technique can be tested, (2) whether it’s been subject to peer review, (3) whether standards exist for applying the technique, and (4) the technique’s error rate.

The above sounds reasonable, but Daubert created an uproar, because “the dirty little secret” of much so-called expert testimony was this: though it was possible in principle to test and validate most forensic techniques, in many cases no one had ever done so. In 2002 one judge even restricted testimony based on fingerprint analysis, saying he was unconvinced the technique was a science rather than a mix of craft and guesswork.


No forensic technique has taken more hits than handwriting analysis. In one particularly devastating federal ruling, United States v. Saelee (2001), the court noted that forensic handwriting analysis techniques had seldom been tested, and that what testing had been done “raises serious questions about the reliability of methods currently in use.” The experts were frequently wrong – in one test “the true positive accuracy rate of laypersons was the same as that of handwriting examiners; both groups were correct 52 percent of the time.” The most basic principles of handwriting analysis – for example, that everyone’s handwriting is unique – had never been demonstrated. “The technique of comparing known writings with questioned documents appears to be entirely subjective and entirely lacking in controlling standards,” the court wrote. Testimony by the government’s handwriting expert was ruled inadmissible.

Prosecutors scrambling to find scientific validation for handwriting analysis last year touted a study by Sargur Srihari, a professor of computer science at the State University of New York at Buffalo. Srihari subjected 1,500 writing samples to computer analysis. Conclusion: In 96 percent of cases, the writer of a sample could be positively identified based on quantitative features of his handwriting such as letter dimensions and pen pressure. Skeptics objected that lab results using a computer prove nothing about what a human can do in the real world, and who can argue? If expert testimony is going to send people up the river, it better be more than someone’s prejudices dressed up as science.

Craig M. Cooley[18] argues that there is a definite need for reform when dealing with the evidence of forensic experts.  He points out that the golden thread which runs through the web of administration of justice in criminal cases is that if two views are possible on the evidence, one pointing to the guilt of the accused and the other to his innocence, the view which is favourable to the accused should be adopted.  Skilled witnesses come with such bias in their minds to support the case in which they are embarked that hardly any weight should be given to their evidence.  A bald assertion by the expert that his deduction is premised upon well-recognized scientific principles is inadequate to establish its admissibility if the witness’s application of these principles is untested and lacks indicia of acceptability

Forensic science has rarely been subjected to the kind of scrutiny and independent verification applied to other fields of applied and medical science. Instead, analysts testifying in courts about fingerprint analysis, bite marks, handwriting comparisons and the like have often argued that in their field the courtroom itself provided the test.

Everyone connected with the justice system expects forensic science to be infallible. It is not.  “Forensic science, including DNA testing, is operating in a noman’s land where there are no accredited standards for the laboratories.

He notes that one of the cornerstones of forensic science is the presumed validity and reliability of scientific test results and interpretations. This presumption must now be viewed through a more cautious lens since the aforementioned cases represent only a few of the ever-growing number where erroneous (honest mistakes) or fraudulent (purposeful errors) forensic science has led to miscarriages of justice.  In particular, the forensic identification sciences (e.g., handwriting analysis, firearms identification, hair analysis, ect.) are primarily responsible for the bulk of these injustices.  According to Barry Scheck and Peter Neufeld, out of eighty-one wrongful convictions fifty-three percent were primarily the result of erroneous forensic science. Moreover, twenty-five percent of the erroneous convictions were attributed to fraudulent and/or tainted evidence. More recently, Saks et al. assert that forensic science errors were a contributing factor in at least sixty-six percent of the erroneous convictions they identified. Additionally, fraud or tainting of evidence was a contributory factor in at least thirty one percent of the wrongful convictions identified by them. This research suggests that forensic science may exceed eyewitness inaccuracy as the foremost cause of wrongful convictions. Consequently, Justice Brennan’s comment in United States v. Wade that the vagaries of eyewitness identification are well known; the annals of criminal law are rife with instances of mistaken identification can now be extrapolated to the forensic sciences. Likewise, with these statistics it could be reasonably argued that from the viewpoint of conventional science, the forensic identification sciences are contenders for being the shoddiest science offered to the courts.


It is submitted that the evidence of handwriting experts in South Africa is characterised by the same deficiencies as in the American criminal justice system.  In essence the problem of reliability of the evidence and the adequacy of proof goes to the heart of the problem.  This is also clear from the submission of Mr Greenfield.  However, in terms of the existing law in South Africa, evidence relating to disputed documents can be given by handwriting experts as well as lay persons.  The adequacy of and weight attached to such evidence is, however, decided by the court.

In view of the outline above it is submitted that the proposals submitted by Mr Greenfield will not solve the identified problems.  Unlike the evidence of finger print experts, the evidence of handwriting experts is not accepted by our courts without reservation as providing prima facie proof of the identity of the writer of a disputed document.  In terms of the case law the acceptance or rejection of such evidence is the prerogative of the court and the court is called upon to determine the reliability thereof in the light of the evidence given.  In all cases the court must be able to justify its decision and to provide sufficient reasoning for its decision.   It is true that the lack of standards to which such evidence must adhere, the lack of a professional body controlling the accreditation of experts in the field and the admissibility of evidence by so-called “experts” of dubious character discredits the profession and casts doubt on the reliability of handwriting experts in general.  It is, however, submitted that the proposal of Mr Greenfield that an ombudsman comprising of three persons is to decide who is to give evidence in court, is not acceptable.  In the first instance, unless good grounds exist for excluding the evidence of a particular witness such decision cannot be taken by an ombudsman on behalf of the court.  The ultimate decision rests with the court and in view with the peculiar problems regarding the reliability of such evidence (outlined in the reference to position in the USA) it is submitted that the proposal cannot be justified.

Furthermore, in view of the questions raised whether or not the conclusions reached by handwriting expert are the result of scientific processes it is submitted that it would be premature to make provision for the acceptance of such evidence in legislation.  The establishment of a professional body for such experts and the accompanying procedural prescripts, the accreditation requirement and the establishment of a code of ethics could go a long way in improving the bad image of handwriting experts and could assist in providing motivation for the acceptance of evidence.  It is, however, submitted that the establishment of a professional body and the accompanying prescripts and procedures is something which must be considered by the profession itself and is not something to be prescribed in legislation.  The intervention by the Department of Justice would not in itself provide more credibility to the process as envisaged by Mr Greenfield.

In the USA it is argued that traditional forensic fields such as fingerprints, questioned documents, toolmark identification, handwriting identification and hair/fiber analysis are not based on science, but rather on subjective comparisons by individual examiners. While physical evidence by itself cannot be erroneous, its subjective interpretation can produce errors, which in turn can result in the ultimate injustice.  In the USA there is substantial support for the reform of structural and regulatory framework of the forensic science community.  It is argued that structural and regulatory inadequacies exist in the forensic science community. The Paul Coverdell National Forensic Sciences Improvement Act of 2000, attempted to address these inadequacies.  It is still argued that while the Act presents a significant step in the right direction, it still fails to address (a) the ‘conflicts of interest’ that exist within the forensic community and (b) the regulatory inadequacies of the community.

As currently organized and structured, the forensic science community in the USA and, possibly also in South Africa, is bound to generate more erroneous, fraudulent and shoddy science, which in turn will produce further erroneous convictions.  As one report in the USA noted, many forensic scientists face a critical deficiency of essential reference material and collections required to perform their jobs. Educationally, funding for graduate-level forensic science research is currently nonexistent.  Similarly, lack of funding has led to “chronic understaffing” in many of the Nation’s crime labs.  In terms of quality assurance, criminal forensics laboratories are a generation behind food, drug, medical and industrial laboratories in their quality assurance systems. Likewise, at present, there are no national standards ensuring the competency of laboratory examiners. Adding to the problem is that in many cases, labs are managed by law-enforcement officials who have no experience with science.  In regard to laboratory standards, forensic science is virtually unregulated… with the paradoxical result that clinical laboratories must meet higher standards to be allowed to diagnose strep throat than forensic labs must meet to put a defendant on death row.  Fewer than half of the 400 to 450 forensics labs in the USA are accredited by the American Society of Crime Laboratory Directors [ASCLAD]. Only New York State requires such accreditation.  Most importantly, too many forensic experts in the criminal justice system manifest a police-prosecution bias, a willingness to shade or distort opinions to support the state’s case.

Regulation of the forensic science community in South Africa can be considered as a solution to the problem, but it is submitted that it should not be done without a proper and more detailed investigation as to the need for reform in South Africa.

[1]   1940 TPD 142 at 147-147; approved in S v Adams 1983 (2) SA 577 (A) at 586 A.

[2]   DT Zeffert, AP Paizes, A St Q Skeen The South African Law Of Evidence Butterworths 2003 on 307.

[3]   A similar provision for civil cases is contained in section 4 of the Civil Proceedings Evidence Act, 25 of 1965.

[4]   S v Boesak 2000 (3) SA 381 (SCA)  at 399D.  

[5]   1941 OPD 33.  See also Stewarts and Lloyds of SA Ltd v Croyden Engineering and Mining Suppliers (Pty) Ltd  1979 (1) SA 1018 (W).

[6]   2000 (3) SA 381 (SCA) 399C.

[7]   S v Boesak 2001 (1) SA 912 (CC).

[8]   1998 (2) SACR 363 (W).

[9]   Science and nonscience in the courts: Daubert meets Handwriting identification expertise Iowa Law Review 82 IALR 21 October, 1996,

[10]     220 F. Supp.2d 548 (S.D. W.Va., Sept. 11, 2002).

[11]    See Fed. R. Evid. 702; Daubert, 509 U.S. at 579.

[12]    See Kumho, 526 U.S. at 147; Daubert 509 U.S. at 589.

[13]    Kumho, 526 U.S. at 14 (quoting Daubert, 509 U.S. at 592).

[14] Science and nonscience in the courts : Daubert meets handwriting identification expertise Iowa Law Review  82 IALR 21 October, 1996,

[15]    Supra.

[16]   The Handwriting on the Wall: The State of Handwriting Expert Testimony in Tennessee Published in the Tennessee Bar Journal, Vol. 34, No. 5, September/October, 1998.

[17] Cecil Adams Is handwriting analysis legit science? 18-Apr-2003  18-Apr-2003.

[18]    Forensic Science and the Death Penalty: Why Reform is Needed in the Forensic Science Community to Ensure that the Innocent Are Not Erroneously Sentenced to Death


Leave a Reply

Your email address will not be published. Required fields are marked *

---- //]]>