January 25, 2010

Criticism of palaeography

I've been working on Thomas More recently, re-reading some of the standard works on its authorship attribution, and there are a few things that have really stood out for me in terms of how we "read" evidence of the sort that has become essential, in Thomas More, to Shakespeare Studies as a whole.

Firstly, I'm entirely happy to believe that the additional passage designated by Greg as in the writing of "Hand D" are a genuine Shakespeare autograph. However, I want to emphasise that word "believe". Because it strikes me that, when it comes to palaeographical evidence, this case necessarily hinges on a willingness to accept a possibility, rather than anything approaching fact.

This was brought home to me by my attempts to read E.Maunde Thompson's "The Handwriting of the Three Pages Attributed to Shakespeare Compared with his Signatures" in Alfred W. Pollard (ed.)., Shakespeare's Hand in the Play of Sir Thomas More (CUP, 1923, pp. 57-112). Now, despite having taken an introductory course in Elizabethan palaeography, I am no expert in handwriting. Even were I, however, Maunde Thompson's article is dense, detailed and highly informed. One of the most skilled palaeographers of his day - perhaps of all time -, his results demonstrated an authority and certainty that are entirely convincing.

The problem here, as I see it, is that too few Shakespearean critics are expert enough in palaeography to actually mount a serious and informed challenge to Thompson's results. There have been a few, but Thompson's double-barrelled argument (this essay following an earlier 1916 piece) lent to the New Bibliographers intent on proving Shakespeare's hand the necessary technical palaeographic support for their other searches. Once canonised in Pollard's volume, Thompson's argument became dogma, and it is still his case which is referred to today, ie: for the palaeographic argument for Shakespeare as Hand D, see Thompson in Pollard (1923).

I can't challenge Thompson. I doubt many Shakespearean critics can. I read Thompson's results, his observations on letter forms, unique curls and particular types of flourish, and I simply have to trust that he's giving an unbiased, objective account of the similarities. Yet I'm simultaneously aware that Thompson's essay is part of an edited collection designed explicitly to bring together the most authoritative voices in Shakespearean criticism in order to consolidate a case for Hand D being a Shakespeare autograph. The agenda of this book is to prove, beyond reasonable doubt, that Hand D is Shakespeare. This makes me uneasy: I cannot objectively assess the validity of the volume's key piece of evidence. Further, when looking at the other essays in the volume, I find Greg and Pollard's historical theories somewhat out of date (based, understandably, on less evidence than we now have), R.W. Chambers' account of parallels of idea extremely unconvincing, and John Dover Wilson's bibliographic links fascinating but of questionable value when negative checks are not performed. Thompson's essay, thus, remains the key piece of unquestionable evidence, and my critical mind objects strongly to my inability to assess its strength.

This doesn't mean that no-one has criticised Thompson, of course. What it means is that very few people have questioned him on his own terms. Instead of challenging the specifics of his parallels and links, sceptical critics have instead cast doubt on the fundamental premise of his work. To wit: how can six extant signatures constitute a sample of Shakespeare's handwriting of acceptable enough size to extrapolate information about his handwriting habits? Particularly when several of those signatures date some ten-to-twenty years (itself a hotly contested matter) after the supposed date of Hand D's contribution. Can a palaeographic argument be considered to have any worth when the 'control' is so negligible?

In 1987, Scott McMillin reminded us of the argument that Hand D is in fact identical with "Hand C" - the supposed playhouse scribe, assumed to be a mere copyist. This argument has been contested, but never completely dispelled. In 2007, Gerald Downs usefully restated this case with a different slant, arguing that mistakes in Hand D are typically scribal and suggesting that Hand C and Hand D may well be separate, but that their similarities may be explained by the idea that both are (separately) scribal. I don't necessarily agree with Downs - again, I lack the skill to have any objective opinion on palaeographic matters -, but his note of caution seems a timely and reasonable one.

Greg, Pollard and co. achieved their goal. They established Hand D as authoritatively Shakespearean in a way which forestalled future argument and shaped critical opinion of the play. The subsequent work of generations of scholars has, for the most part, agreed with them and used a variety of criteria and methodologies to strengthen the case - though, it has to be said, never unanimously. It's a cumulative case that, as I stated at the beginning of this blog, I am content to buy into, to claim a belief in. But I feel it only responsible to make clear that it is a belief,  based on my willingness to accept a palaeographic argument that I lack the means to confirm, and with due caution that my core texts for this belief were constructed with a deliberate agenda to enforce that belief in less informed minds. I also believe that the Shakespearean attribution has seriously damaged, as well as enabled, scholarship on Thomas More - for, in the flurry of activity to ascertain Shakespeare's authorship (or otherwise), the play itself has remained criminally overlooked, and evidence relating to company and date has been forced to yield if that evidence doesn't neatly comply with Shakespeare's opportunity to contribute.

- 2 comments by 1 or more people Not publicly viewable

  1. Duncan

    I got a distinct chill down my spine when confronted with the Hand D sample when it was on display as part of the Shakespeare exhibition at the National Portrait Gallery a few years back. Your post contains a useful summary of the ongoing attribution debate, which I imagine most people with a casual interest in Shakespeare, like me, are unaware of.

    It strikes me that this must be an area ripe for computer analysis. If computers are trusted to analysis the style of a text, surely something more readily quantifiable like the shape of written characters must be even easier to analyse. The added bonus being that computer programs have no reverence for individual writers and therefore treat handwriting samples objectively! Scholars making subjective judgments surely just see what they want to see?

    25 Jan 2010, 12:35

  2. People have high hopes for computers in this area. However, there are several key problems:

    1) Quality vs. quantity. Computers are very useful in quantitative research, but less so in terms of identifying the quality of a similarity. Compare any two people’s letter ‘a’s and look at the minute differences in how steep the topmost curve is. What data do you give a computer in order to determine how similar they are? What are the empirical defining limits for closeness?

    2) Degradation. Most of the Elizabethan manuscripts we’re concerned with are in various stages of degradation. You have to find a way of turning the physical manuscript into a form which a computer can use and analyse. Conventional scanning will miss many subtleties, and when you’re trying to be this precise, you need to be able to see exactly how long, say, that ‘j’ is. The computer also needs to be able to tell the difference between mistakes, crossings out, letters run into one another, abbreviated forms, marks on the paper, split pages etc. etc. etc., the kind of data which humans can obviously pick up, but which you would have to individually programme into a computer to meet the specific characteristics of each unique MS.

    3) Size of sample. This problem of sample size is always going to be a problem in the case of the More manuscript. We’re lacking samples of so many letters that we know, without question, are written in Shakespeare’s hand. How can we programme a computer to extrapolate probability of other letter forms from the tiny sample available? This is where the weaknesses of the More arguments are exposed: the palaeography argument is necessarily, to an extent, subjective. Plus, as mentioned in the blog, there’s no way to objectively count for the possibilities of handwriting changing over time.

    4) Negative checks. In order for a computer to establish the relative likeness of two pieces of script, it needs a considerable checking sample to compare them against: there’s no point saying two letter ‘t’s are very similar if you then find that half a dozen other writers were using a form so close as to be indistinguishable.

    Computers are one of the most important developments in attribution studies, but apart from in the comparison of facsimiles, I believe they’re only ever going to be partially useful in physical matters such as handwriting. There’s too much human intervention required, too much manipulating of the evidence in order to present it in a form palatable to the machine. Sadly, I think that for the foreseeable future, it’s going to remain a very human, very subjective and very fallible endeavour.

    25 Jan 2010, 13:02

Add a comment

You are not allowed to comment on this entry as it has restricted commenting permissions.


I’m Peter Kirwan, a final year doctoral student in the English Department at Warwick, and this is my PhD blog.

Conferences, reviews, articles, thoughts and links relating to my interests in the Shakespeare apocrypha, early modern drama, authorship and performance.

January 2010

Mo Tu We Th Fr Sa Su
Dec |  Today  | Feb
            1 2 3
4 5 6 7 8 9 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25 26 27 28 29 30 31

Search this blog



Most recent comments

  • Congratulations. by Andrew on this entry
  • Congratulations, Peter! Whenever you feel up to firing a PDF of it my way, I'd love to read it … by Edmund King on this entry
  • Linda – Paul didn't talk on biography, but he did run a seminar on it. I'm afraid I couldn't attend … by on this entry
  • Shakespeare was immersed in a variety of European literary cultures and it seems only fair that thes… by Duncan on this entry
  • Good luck Pete! Hope it all goes well. by Sylvia Morris on this entry

Blog archive

Not signed in
Sign in

Powered by BlogBuilder