Forensic handwriting analysis and Cuneiform tablets
If any of you heard the paper that I gave in Chicago a year ago you may remember this document. It’s a love letter written by a young girl to her boyfriend. The problem was, the girl was 13 when she wrote it, and the boy 19. Three years later she claimed that he had raped her; not just statutory rape, intercourse with a minor under the age of consent, but non-consensual sex, which of course carries a much heavier penalty. The letter turned up, by chance, in the course of the trial; it obviously has considerable bearing on the matter at hand, particularly, crucially, because the girl denied that she wrote it.
The document was sent to me by the lawyers acting for the defence because I am what is known as a Forensic handwriting expert, and have practiced as such for nearly thirty years. What I’m going to do in the next few minutes is go a little more deeply into the methodology of forensic handwriting analysis as practiced nowadays, and then show you an attempted application of it to cuneiform tablets.
When you receive a document like this, you follow a standard procedure. There are various hardware tests you apply, which are not relevant here; but the main task is to build a picture of the handwriting. This is of course alphabetic writing, so there are a possible 26 upper case and 26 lower case letters, possibly a further 26 cursive capitals (as opposed to block capitals), plus numerals and punctuation. Every instance of every one of those items that occurs in the document at hand must be taken into account. The procedure is very laborious, and very thorough.
What you do is, you go through the document letter by letter. In this document the first word is Dear. I take the D, and copy it into a chart, which is laid out alphabetically. I have a capitals chart, a lower case chart, and a numerals and punctuation chart. The next letter is e: that gets copied into my lower case chart against e. But: the next time I come to an e, I make a decision, a rather subjective one: do I already have that kind of e, or is it different to any e that I have already copied? If it is the same as an e that I have already, I ignore it; otherwise, I copy it.
Here is a chart in the process of construction, so that you can see what I mean. So far, for instance, I have found three a’s. In the first one, the vertical straight line on the right only goes a short way down towards the baseline. In the third, it goes all the way. In the middle, it goes a little further than the first, but not all the way. Also, in the first, the bowl of the a (that is, the circle that makes up its body) is not closed; in the other two, it is.
The idea is that by the time I’ve gone carefully through the entire document I should have captured instances of all of the different kinds of every letter represented there. I define difference rather generously, as you can see.
A word on methodology. I used to draw the letters: as it were, ironically, to forge them. Forensic handwriting experts trained in that traditional method would make excellent forgers. I now use Photoshop.
The next stage is to go through the comparison document, doing exactly the same thing, independently of the questioned document.
One ends up with two charts: one is a comprehensive description of the handwriting of the questioned document, the other of the sample document. Then you compare them. To prove identity, I would want to find, in the sample writing, a good match for every item in the questioned writing. I must stress: the effect that one is going for here is of overwhelming probability. People go to prison on the basis of document evidence. The standard of proof for that is, beyond reasonable doubt.
I then make a new chart
that combines both the working charts, to show either difference, if that is my opinion, or identity. Here I am attempting to show identity. Note, I am presenting here only a selection of the evidence. I do one good match for every letter. The effect is still overwhelming, and you don’t want to bore a jury. If in cross-examination the barrister says, but what about this letter here, it looks different, then I have my notes, and in fact I usually have the whole thing in my head anyway, so I can say, very politely and helpfully, yes, that’s very interesting, the leg of the a does go all the way to the baseline, and so it does in this word in the sample writing too. The barrister finds that he or she is making the other side’s case, and very soon abandons that line of argument.
The question is, how does one apply that technique to cuneiform writing? Well, you make a chart.
