What are the essential elements in a script that make it legible? Looking at peoples’ handwriting is a good way to assess what the basic form of letters is understood to be, stripped away of typographic styling. Seeing “in” and “out” strokes, and how the writing tool is used can help you learn to write the script like a native user, and, if you’re designing a typeface, can inform your design decisions.
Visit the great collection complied by the ITF (Indian Type Foundry).
Comments
Safety notice: 1. there is more to type design than just studying handriting. By studying handwriting only, one is running into a risc of ignoring typographic tradition of treating certain characters (in Latin compare handwritten -a- and typographic upright -a- for example) 2. besides the good ones, there are bad handwriting styles out there and these will teach you very little 3. sample of handwriting by itself does not say anything about in- or out-strokes (unfortunately! :). You can only guess. IMHO writing manuals are much better in this respect as they: a. show the stroke order b. are usually carefully picked samples of “good handwriting” (i.e. something somebody very familiar with the writing system considered a standard). It is however, still very useful to look at handwriting and there are lessons to be learned.
Just wanted to add that.
Cheers from CE
A great resource for earlier stages of Gujarati handwriting is in the Avestan Digital Archive:
http://ada.usal.es/paginas/buscador_biblioteca
The archive contains high quality digital images of Zoroastrian Avestan scriptures written in the Avestan script, and most of the documents also contain annotations or commentaries, sometimes pages in length, in either old Nagari or Gujarati script (and sometimes Indo-Arabic script). The Gujarati ranges from scrawls to the most exquisite calligraphy in a couple of cases; the best examples are from the Meherji Rana Library collection.
My impression from studying historical samples of Gujarati script (together with Kaithi and Marwari/Mahajani, which are closely related) is that the main difference between more formal Gujarati handwriting (and typographic Gujarati) and informal varieties is that the latter conserve the older trait of joining cross-strokes to the top of the stem, whereas the former have brought the join back down to a point near the middle of the stem, probably under influence from Devanagari. In some people’s handwriting, the join at the top of the stem is sharp or stopped, whereas in others it glides or curves at the transition.
There is an older form for ‹a›, with three whiskers to its left, that appears both in pre-20th century handwriting and typefaces but has been replaced with the form that begins like ‹r›. Apart from this and the general preference for closed loops (‘box loops’) in formal typefaces, there is little in the way of substantive differences between typeface and handwritten forms that can be compared to idiosyncratic differences in specific letters like those between handwritten versus typeface ‹a› or ‹g› in Latin script.
@Chris: wonderful source this Avestan Digital Archive! I have seen some old manuscripts in the British Library and the excellence of this is nearly there (just missing the dust and smell). Should you have any other sources like this, please cc: me. I would be most grateful.
What about handwritten ‹Ka› vs. typographic ‹Ka›? I have seen that _sometimes_ the Ka is written using a single stroke which leads to a rather different structure (maybe I have mistaken it for other character, that is quite possible). Also vowel marks can connect to base syllables in different manner. In typographic script they do not always connect. It may have been (and probably is) an influence of printing technology, but could be considered an established typographic treatment. But indeed the case of ‹a› and ‹g› in Latin is extreme (also perhaps because the typographic history is longer?). Mentioning “substantive differences” brough one more drawback of contemporary handwriting with agnostic tool (as Gerry Leonidas calls it, I think, i.e. a tool which does not change its trace depending on the orientation [as opposed to a broad-nib pen for example]). Such handwriting does not show the contrast between thicks and thins and the stress axis. And this is of great significance in typography!
@Erin: I will not post any more comments until you change the colour of the text in the comment field (and perhaps elsewhere too) to black.
David-
I’m not exactly sure what you are referring to when you mention the handwritten vs typographic ‹k› shape, but I suspect you mean the contrast between the modern “crossed S” stroke sequence and the older way of writing it.
My interest in this comes from my research on the origins of several now mostly defunct scripts used in Indonesia and the Philippines, which unexpectedly turn out to be directly derivable from the older informal Nagari handwriting that modern Gujarati script derives from (along with Kaithi further to the east and Marwari/Mahajani as well). The modern Gujarati ‹k› (and ‹ph›) is/are a reinterpretation of the “curly Z” shape that comes originally from the way the left side curl (just like for ‹v›) was extended upward to incorporate the location of the original headstroke. This made it awkward to draw the stem vertically to meet the beginning of the tail, hence a pressure to draw it askew, leading to “curly Z”. In Kaithi and Marwari/Mahajani at least, this evolved in some varieties to a “3”-like shape via a shape where the beginning of the “3′ started with a short downward stroke; in other words, the glide point at the bottom of the first curl changed to a sharp, stopped join and the sharp join with the top of the stem changed into a glide point. I haven’t yet seen an actual example of the “3” shape in Gujarati writing as such, but I am sure it must have been a variant just as much as in Kaithi and Mahajani, or even in informal Nagari handwriting: it has to have been there because the evidence is so strong that the Indonesian-Philippine scripts derive from early informal Nagari and the Sumatran scripts’ ‹k› clearly derives from a “3” shape. In any case, the modern standard Gujarati “crossed S” shape is clearly a reinterpretation of the “curly Z” that somewhat simplifies the stroke sequence, at the expense of lifting the pen to transition between the end of the letter’s body and the cross-stroke. As far as I can tell from the limited sources I have been able to find, this stroke sequence probably came into currency in the 19th century. It would probably be useful to compare the dates of the Avestan manuscripts with the letter styles in the Gujarati annotations to figure out when this change took place in handwriting.
Here are sources I have for older Gujarati typestyles:
Faulmann, Carl. 1878. Das Buch der Schrift. Vienna: K. K. Hof- und Staatsdrucker. (available at Google Books)
——————– 1880. Illustrirte [sic] Geschichte der Schrift. Vienna/Pest/Leipzig: Hartleben.
Ramsay, H. N.. 1842. The Principles of Gujarati Grammar. Bombay: Imperial Press. (available at Google Books) (This source alternates unpredictably and chaotically between different variants for the same letter in Gujarati script and even their Devanagari equivalents, sometimes including different types within a single word. You will only understand what is going on if you have already absorbed the older variants given in Faulmann.)
And here are some extra sources for Kaithi/Kayathi, Mahajani and handwritten Devanagari:
Grierson, George A. 1903-1916 (2005). The Linguistic Survey of India, 11 Vols. in 19 Parts. Delhi, Low Price Publ. (2005) ISBN 8175363614 (A massive work with numerous reproductions of handwritten specimens from various scripts, in particular Kaithi and related scripts)
————————. 1881. A Handbook to the Kayathi Character. Calcutta: Thacker, Spink, and Co. (available at Google Books)
————————. 1899. A Handbook to the Kaithi Character. 2nd rev. ed. of the title A Kaithi Handbook, 1881. Calcutta: Thacker, Spink & Co.
Kellogg, S. H. 1875 (revised 1893, 1938). A Grammar of the Hindi Language. New Delhi: Oriental Books Reprint Corporation (1975 Indian edition).
Palmer, Edward H. (n.d.). Oriental Penmanship. London: Sampson Low, Marston & Company. (available at Google Books) (Samples of Devanagari of different degrees of formality: the informal styles share some important features with Kaithi and Gujarati)
Pandey, Anshuman. 2007. Proposal to Encode the Kaithi Script in ISO/IEC 10646. http://std.dkuug.dk/jtc1/sc2/wg2/docs/n3389.pdf (Very well documented propsal for encoding Kaithi in Unicode)
————————-. 2010. Preliminary Proposal to Encode the Mahajani script in ISO/IEC 10646. http://anshumanpandey.com/
Palmer’s _Oriental Penmanship_ is very interesting for the several pages on Devanagari handwriting at the end. In the most informal style illustrated, you can see three different ‹k› variants cooccurring in the same text: “curly Z”, “3”, and the intermediate form.
If anything I wrote above is unclear, you can contact me at christophermiller@mac.com and I can send you an illustration of what I mean.
Great discussion, you guys. Thank you!! Now I have to catch up and check out all of those links. Keep ’em coming!
Another link you might find interesting for a view of the range of historical variation:
http://www.informaworld.com/smpp/content~content=a907362438~db=all~jumptype=rss
This is an article (with illustrations of the script) about a 1750 Gujarati nautical map of the Bay of Cambay.