Forensic handwriting analysis

DEFINITION: Examination of samples of handwriting, usually to establish validity, fraud, or forgery.

SIGNIFICANCE: Forensic experts who examine handwritten as well as typed materials assist in law-enforcement investigations by comparing handwriting or signatures, analyzing disputed documentation, and examining and identifying legal documents. These experts, who often specialize in particular kinds of writings, such as real estate records, wills, or medical and dental records, are often called upon to present testimony on their findings in court.

The field of questioned document analysis, of which handwriting analysis is a subset, is one of the oldest disciplines in the field of forensic sciences. The practice of forgery is as old as written communication. Under the Code of Justinian in 539 CE, Roman law contained rules for identifying and comparing handwriting. A questioned document is any document whose authenticity is questioned, either entirely or partially. Questioned documents can include concert tickets, postage stamps, bus passes, and dollar bills. Expert handwriting examiners are concerned primarily with anything written by hand in questioned documents; they apply the basic rules of document analysis and their own experience to distinguish real writing from forged writing.

During forensic handwriting analysis, samples of signatures may be compared, alterations and obliterations of documents may be scrutinized, and ink may be subjected to microscopic analysis. Document examiners may be asked to apply analytical techniques to documents in criminal investigations or to testify in disputes regarding written threats, unsigned letters, attempts to extort or defraud, contract disagreements, or identity theft. Many document examiners work for government agencies; others are in private practice, often providing their services to investigations of medical malpractice or insurance fraud. A may be called upon to verify a person’s signature on a sign-in sheet to prove or disprove that person’s claim to have been at a particular place at a particular time. Document examiners also determine who signed checks, credit card invoices, and contracts in investigations of fraudulent activities. By identifying who added an addendum to a victim’s will, a document examiner may free the spouse from suspicion.

In analyzing handwriting, document examiners determine only the physical characteristics of the handwriting. Traits of the writers’ personalities are not addressed, nor can document examiners ascertain from handwriting such other aspects of the writers as age, sex, race, or educational status.

Handwriting Development

In the United States, children have traditionally been trained in whatever handwriting style is in favor when they are in elementary school. Depending on geographic location, children might be taught one of a variety of copybook styles of penmanship, such as Palmer or Zaner-Bloser. In the early days of learning to write, children focus attention on how to write the letters of the alphabet correctly. As schooling continues, these methods become increasingly automatic, and individuals’ writing begins to deviate from the standards originally learned as the focus of writing shifts toward what is being written and away from how it is written.

As the brain matures, an individual’s neuromuscular coordination and visual perceptions become unique to that person. By the late teenage years, a person’s handwriting has matured, but it has not stopped changing. As people progress through life and even through various emotional or physical upheavals, their handwriting changes as well, and sometimes not in subtle ways. Despite such long-term variations, a person’s handwriting retains features that identify its author.

Collecting Samples

Before comparisons can be done, the handwriting examiner must have samples of known, standard writing—unquestioned samples—from the individual of interest. Good standards are essential for valid results.

As much as possible, the conditions under which the known standards are produced should duplicate the conditions under which the questioned material was written—with the same or similar writing instrument, in the same writing position, on the same kind of surface, and on the same type of paper (lined or not, the same kind of stock). If the questioned material was written in cursive, the known standards must be written in cursive.

At least one of two kinds of writing standards are obtained: nonrequest and requested writing. Nonrequest writing is also known as spontaneous or undictated writing, and samples of requested writing are also known as dictated exemplars. Nonrequest writing is usually material written previously by the individual; such material is much more likely than requested writing to reveal the individual’s normal, unforced writing habits. As the writer was ignorant of the way the writing would eventually be used, conscious or unconscious attempts to alter, disguise, or even improve the writing are lacking. Nonrequest writing may be difficult to get authenticated for court, however. In addition, the examiner may not be able to obtain enough nonrequest writing for good comparison, and such material may have been written under conditions different from those of the questioned material.

Requested exemplars are standards produced by request, usually in the presence of one or more witnesses. Using such exemplars can be advantageous in that they allow for better duplication of the conditions under which the questioned material was written, and authentication for court is often easier. Collecting requested exemplars, however, calls attention to the writing process, and regardless of innocence, most writers tense up, changing their handwriting subtly or not so subtly.

Most examiners prefer to receive both requested and nonrequest writing standards for comparison to questioned material. Nonrequest writing standards can be obtained from many places, including job applications, bank records, business and real estate agreements, employment records, tax returns, wills, and letters. Most of these kinds of sources, however, provide signatures only, and signatures are not well suited for handwriting comparisons because they are of insufficient length and often do not bear much resemblance to the rest of their writers’ handwriting.

Comparison

Common features of handwriting, such as slant and letter formation, are class characteristics—attributes learned in grade school. Uncommon handwriting features are individual characteristics that are distinctive and peculiar to one person’s writing. Document examiners look beyond class characteristics as they use both vision and microscopy to compare questioned handwriting or signatures to known standards. Experienced examiners recognize common traits and refrain from making identifications from these; instead, examiners rely chiefly on individual characteristics.

Examiners inspect handwriting for minute details, looking at traits such as letter size, how letters are connected, how beginning and ending strokes are formed, the height ratio of capital to small letters, spacing between letters, words, and lines, spacing on the page, the skill level of the writing, and the speed and pressure of the writing instrument on the page. They also observe execution—in general, does the writing in the known sample look natural, or does it show characteristics lacking in the original? Experienced examiners can discern patterns in the consistent combination of less-common traits with more-common ones. When the questioned sample begins to diverge too much from the standards, it is likely false.

No two people write exactly alike. It is also true, however, that no one person writes identically twice. People hold their pens or pencils at varying angles, and they write at different speeds at different times. In one person’s writing, not every instance of any letter will be identical; beginning and terminal strokes will vary. These variations in one person’s handwriting, however, still constitute a range that is recognizable to experienced examiners. For example, in the letter combination "th," some people tend to make the "t" or the "h" higher, and some usually make the two letters of equal height. When such details vary within an individual’s known standards, a skilled examiner notes the differences among the standards and compares the pattern with the questioned material. In a study of two hundred people, it was found that 5.5 percent made the "t" taller than the "h," 78 percent made the "t" shorter than the "h," 15 percent showed no pattern, and 1.5 percent made the "t" and the "h" equally high. Such seemingly trivial (and sometimes tedious) details are carefully noted by handwriting examiners.

Forgery Types

Generally, forgeries are freehand simulations, tracings, or normal-hand simulations. For freehand simulations to be done well, correct letter formations and height ratios must be written at the same speed and with the same pen pressure as in the original, all at once. At the same time, the forger must suppress his or her own natural individual writing traits, which can be very difficult. If the forger concentrates on making letter formations like those in the original, the pen pressure will likely be different, and the speed will almost always be much slower than the original. If the forger concentrates instead on speed and pressure, letter formations and connecting strokes revert to the forger’s.

Even if the forger practices the to-be-forged writing, it is likely that very subtle details, such as line quality and placement of pen lifts, will diverge from the original. A signature that, when genuine, is illegible may suddenly, when forged, become legible, or it may appear hesitant or drawn instead of dashed off in a well-practiced manner. Microscopic examination of such a forged signature reveals "patching"—the retouching of a line or blunt starts and stops indicative of slow, concentrated copying. Examined competently, freehand simulations seldom have much chance of deceiving the examiner.

Forgeries involving tracings have their own shortcomings. A forger can use carbon paper or graphite to copy an original signature or handwriting, but remnants of the carbon paper or graphite are often microscopically detectable. A forger might place a page over the sample or signature and trace it, but when tracing is done atop the original, the line quality of the tracing is generally poor. Where the original appeared smooth and fast, the tracing appears slow, its lines wavy and drawn. Instead of being relatively consistent, the pen pressure varies, and natural differences in line shading disappear. If a forger makes an impression of a signature or handwriting with a pointed object and then uses a pen to fill in the impression, the indent in the paper is detectable. If the original or a copy of the original writing is available and the writing being examined mirrors it exactly, it had to be traced or somehow reproduced; no two handwritten samples are identical.

In normal-hand forgery, forgers write normally or change their writing so that they can later deny authorship. To disguise their writing, forgers may change the slant or size of it, add strokes to letters, or alternate between uppercase and lowercase letters. It is relatively easy for individuals to disguise their handwriting, at least to a degree that may not be detected by the uninformed, but it is difficult to maintain such changes as the writers’ own tendencies ceaselessly assert themselves. Writing is a well-learned process, and years of practice leave their mark.

Legal Challenges

Handwriting examination is not an exact science. The methods used by different experts vary, and training techniques differ as well. In 1989, a lawyer published an article that criticized the use of expert handwriting examiners in court, stating that was lacking to prove that document examiners perform any better than nonexperts and that the error rate of document examiners was high. Challenges to the validity of forensic handwriting analysis ensued, and one court ruled handwriting analysis inadmissible, calling the comparison of known writings with questioned ones entirely subjective and lacking standards.

Evidence to the contrary appeared, however. In a 1997 study of experts and nonexperts asked to identify handwriting, the error rate for the experts was 6.5 percent, whereas that for the nonexperts was 38.3 percent. The nonexperts had incorrectly matched documents made by different writers at almost six times the rate of the experts. In a 2001 study, the error rate of document examiners in determining false signatures was found to be less than 1 percent (0.49 percent), whereas that of nonexperts was 6.47 percent. In the same study, the rate at which experts declared genuine signatures false was 7.05 percent, and the error rate for nonexperts was 26.1 percent. A professor of computer science at a New York university reported that the writer of a sample could be determined correctly in 96 percent of fifteen hundred samples scanned into a computer programmed to measure such aspects as letter dimensions and pen pressure. In 1999, the US Court of Appeals found that the lawyer who in 1989 questioned the overall reliability of handwriting examinations had no standing to do so because he lacked training or education in the field, and handwriting analysis was again admissible in the courts.

Bibliography

Bauchner, Elizabeth, editor. Solving Crimes with Science: Forensics Document Analysis. Mason Crest, 2014.

Ellen, David. Scientific Examination of Documents: Methods and Techniques. 3rd ed., CRC Press, 2006.

James, Stuart H., and Jon J. Nordby, editors. Forensic Science: An Introduction to Scientific and Investigative Techniques. 2nd ed. CRC Press, 2005.

Kelly, Jan Seaman, and Brian S. Lindblom, editorss. Scientific Examination of Questioned Documents. 2nd ed., CRC Press, 2006.

Lewis, Jane A. Forensic Document Examination: Fundamentals and Current Trends. Academic, 2014.

Morris, Ron N. Forensic Handwriting Identification: Fundamental Concepts and Principles. Academic Press, 2000.

Slyter, Steven A. Forensic Signature Examination. Charles C Thomas, 1996.

Songer, Mark. "Forensic Handwriting Analysis." Robson Forensic, 16 Nov. 2023, www.robsonforensic.com/articles/forensic-handwriting-analysis-expert. Accessed 15 Aug. 2024.

Willingham, Emily. "Forensic Experts Are Surprisingly Good at Telling whether Two Writing Samples Match." Scientific American, 2 Aug. 2022, www.scientificamerican.com/article/forensic-experts-are-surprisingly-good-at-telling-whether-two-writing-samples-match. Accessed 15 Aug. 2024.