Questions & Answers

The author of this site is Andrew Casson. I'm a professional software engineer, a semi-professional musician specializing in the medieval repertoire, and a keen amateur philologist. So when I had both the idea and the opportunity to create this project around the Cantigas de Santa Maria, it pushed all my buttons, so to speak.

Languages, and their history, have fascinated me for as long as I can remember, probably going back to when I was five years old, and was taught to count to ten in French (in an appallingly bad accent) by one of the boys from the “big school” on the school bus. As a (rather long-term) consequence, I very nearly embarked on a Ph.D. in computer-assisted historical linguistics at the University of Sussex, when fate intervened and I ended up with a job at AIAI in the University of Edinburgh in late 1990, at the age of 24.

My academic and commercial software engineering career from that point—which ended up in the role of Principal Engineer developing pre-press applications for a well-known Yellow Pages publisher—probably won't be of much interest to readers of this website. What Edinburgh really did for me was introduce me to medieval and renaissance music, initially via the route of historical re-enactment activities, but eventually in a semi-professional capacity as a member of Chris Elmes' group Gaïta. These days my personal repertoire covers a wide range of European music from the ninth to the sixteenth centuries, but at its core are the early instrumental dances (saltarelli, istampitte and estampies), French, Italian and English Renaissance dances from the 15th and 16th centuries and, of course, the 13th century Cantigas de Santa Maria.

Having left Edinburgh in the Summer of 2010, I now live in the green and cloudy north of Gran Canaria with my partner Marianne, but I'm pleased to have been able to continue my collaboration with Chris Elmes in bringing the Cantigas de Santa Maria to a wider range of performers, his focus being on the music and mine on the words.

Two answers, one practical, one more philosophical. The first is simply that, as far as I could tell when I started, nobody had done the work to actually regularize the metre of every single line of every one of the Cantigas de Santa Maria before, or at least if they had, no results were published that would be of any use to performers. So I saw a gap, and decided it should be filled. The second answer is that, whilst the Cantigas have obviously been the subject of much recent musicological, philological and paleographical study, it's very easy to lose track of that fact that they weren't written to be studied, but to be sung. So whilst I do naturally hope that some academic merit will be recognized in the work that I've done, it's far more important to me that the Cantigas be performed, recorded and enjoyed all the more as a result.

The core data from which this website is generated is a set of 427 text files, one for each Cantiga. These are based on transcriptions taken from Mettmann's three-volume 1986–1989 edition of the Cantigas de Santa Maria, but with a variety of custom-built mark-up, which allows me to indicate places where metrical processes such as synalepha, synaeresis and diaeresis are required. The mark-up also lets me make non-destructive changes to the original text, either for the purposes of making basic corrections, or to introduce further textual modifications where necessary (such as complete elision of vowels) to improve the scansion. The file format is carefully designed so that both Mettmann's original transcriptions and my own published text can be generated from a single, unified source, as well as allowing compilation of all of the lists and statistics that you can find on the corrections and modifications pages. The text files also contain my own footnotes, again with mark-up that allows the generation of cross-references between cantigas, links to the concordance, and so on.

The software that I have written then takes these input files, and applies generic syllabation rules for medieval Galician-Portuguese, along with any special metrical mark-up for each line, in order to generate the bulleted text and syllable counts that you see in the HTML pages delivered to the web. The software also automatically applies my own spelling changes on output, and uses TTS (text-to-speech) rules to generate the IPA transcriptions and work out the rhyming schemes and other metrical data. Finally, it builds the SQLite database used by the concordance search.

That's basically it, except for a few bells and whistles. Of course, in reality, I didn't just do all the text mark-up in one go, then run the software once. It has been a very long and arduous process of refining the input files and the software in parallel, through countless iterations. By the way, in case anybody is wondering, the software is written in Perl, which absolutely wipes the floor with every other programming language on the planet when it comes to this kind of thing.

For the early stages of the project, I have ensured that my texts are compatible with the edition of the Cantiga music produced by Chris Elmes and published by Gaïta Medieval Music. This is probably the only full edition of the music available at the moment which is both reasonably priced and easily acquired, and you can order it from Gaïta's publications page. Please note though that the fourth (and final) volume, containing Cantigas 301–427, has not yet been published, but is planned for 2012.

Other editions of the Cantiga music are of course available, but they are very unlikely to be found in your local bookshop, or even on the web: you'll probably have to go to a (large) library. The ones you are most likely to come across are La música de las Cantigas de Santa María del Rey Alfonso El Sabio by Higinio Anglés (3 volumes, Barcelona 1943–1964), and the more recent work Cantigas de Santa María: nueva transcripción integral de su música según la métrica latina by Roberto Pla (Madrid, 2001).

The more adventurous performer can, of course, go straight to the manuscripts. Greg Lindahl's excellent Cantigas de Santa Maria website has an (almost) full set of online facsimiles of the [E] and [To] manuscripts, and I have linked these directly from the individual cantigas here (under the floating 'M' button on the right of each page). Don't expect to get very far with the musical notation if you haven't seen it before though.

However... the second stage of my project—which has already begun and is making very good progress—will be to produce full performance scores of every cantiga, with the words of every stanza underlaid directly to Elmes' music. These will take the hard work out of preparing the cantigas for performance once and for all, and will, I hope, become a new standard reference for performers of early music. The scores will, of course, be available from this site as soon as they are published. Watch this space!

For a brief narrative summary in English of most of the Cantigas, you can now consult the Oxford CSM Database: just click on a cantiga in the Poems tab and you'll find the summary there, along with other useful information, although not the full original text. As of September 2011, however, there seem to be no summaries of the Cantigas de Loor (numbers 10, 20, 30 and so on), presumably because these don't have stories as such.

Proper translations of the Cantigas de Santa Maria are not that easy to come by, but you might get lucky in a good library or specialist second-hand bookshop. The only complete edition in English that I know of is:

Songs of Holy Mary of Alfonso X, the Wise: A translation of the Cantigas de Santa Maria
Kathleen Kulp-Hill
Tempe, Arizona, 2000.

but as of June 2011, this is out of print. In Spanish, the only edition I know of is:

Alfonso X, el sabio: Cantigas de Santa Maria
José Filgueira Valverde
Castalia, Madrid, 1985.

but this contains only the 200 or so Cantigas found in the [T] manuscript. I don't know for certain that this is out of print, but I strongly suspect so; please contact me if you know otherwise.

(Note, by the way, that neither of these two works includes transcriptions of the original Galician-Portuguese text or the music, so if you are fortunate enough to locate affordable copies, don't buy them with that expectation.)

However, anybody with a working knowledge of modern Spanish, Portuguese or Galician should be able to make reasonable sense of the language of the Cantigas, and this will improve over time, of course. I would just like to finish off this answer with a plea though: don't perform anything you don't understand! Not only will knowing what you're singing about improve the delivery immeasurably, but it's always better to be in a position to judge the appropriateness of the content for your audience, even if you think they're not going to know any better. It's an unavoidable fact, for instance, that many of the Cantigas are sprinkled with the kind of casual anti-Semitic and anti-Islamic sentiment that was typical of the medieval period. If you're concerned about this, do your best to find translations of what you're planning to perform, and make sure that you at least put it into the appropriate historical context for your audience.

The two main things that I have planned for the future are:

• finishing marking of the open or close pronunciation of the vowels e and o in the main texts, i.e. [ɛ] vs. [e] and [ɔ] vs. [o] (see the Pronunciation guide for more on this issue); and
• full performance scores of every cantiga, in collaboration with Chris Elmes (see Where is the music? above).

Please let me know if you have other ideas for things you'd like to see here!

It would depend on what the project involved, of course, but I would certainly be interested to know more, whether you're a performer, an academic, or indeed anybody else with a genuine interest in the Cantigas. Email me if you have a proposal.

	Casson, Andrew D. "Cantigas de Santa Maria for Singers". Web, 21st June 2011.
	<http://www.cantigasdesantamaria.com/>

Substitute the specific date when you access the site, since pages may change over the course of time. I would also appreciate it if you would drop me a note when you cite me. Thank you.

Sorry, but no, you can't. You may not republish my work in any form whatsoever.

Basically, the principle here is credit where credit is due, which I hope nobody will take offence at. Although my own work builds upon the exemplary work of Walter Mettmann, it has so far occupied a good solid year of personal full-time effort—marking up and correcting transcriptions, poring over manuscript facsimiles, proofreading, problem-solving, research, discussions with musical colleagues, web design, programming (getting on for 10,000 lines of code), testing, and so on. I simply don't want copies of my work appearing without my name on them, neither do I want out-of-date versions of my work, or versions modified by somebody else, to appear with or without my name on them. The only way to avoid both is not to allow any copies to be made, and this is the only thing I ask in return for making this site available freely to all.

Even if these existing online transcriptions were of good quality, you'd still need to have a solid understanding of the medieval Galician-Portuguese language as used in the Cantigas, and you'd then need to duplicate all the work I have done in order to produce something that is ready to be performed in full. Unfortunately, however, all copies of these online texts that I have found so far—and there are quite a few around the web, though they clearly share a common origin—are of very poor quality, in fact so poor as to be unusable for performance. To begin with, they are riddled with dozens of OCR errors—such as "Sermor" for Sennor, "ab" for ali, "magãa" for maçãa, and absurdities such as the word "apelidoso" where the original word is apelido and the "-so" is actually Mettmann's line number 50. Much worse, though, is that fact that every single instance of the letters ẽ, ĩ, ũ and ỹ (that's e, i, u and y with tilde) has been dropped completely. That's a total of 2620 missing letters. Not only does this result in nonsense, but these four letters always represent a separate syllable, and their loss means that there is absolutely no chance of reconstructing the original metre and fitting the words correctly to the music. (A fine example of this is the mangled text "Ond' ha gran cuba cha devo pararon tal" from CSM 351:23, which should in fact read Ond' hũa gran cuba chẽa de vỹo pararon tal in Mettmann's original transcription.)

Simple question, rather technical answer: the r in the unelided form que•ro begins a syllable, and elision doesn't change that fact. The elided syllables must therefore be que and reu, not quer and eu. Whilst this makes no practical difference with the consonant r, it certainly does with n and s. Take CSM 31:31, where the word dano is elided:

sen da•n' e sen o•ca•jon

This contains syllables da and ne, so the n gets its normal syllable-initial pronunciation, resulting correctly in [da·ne]. Division as dan and e would make the n syllable-final, and imply pronunciation as [dã.e] or [daŋ.e] which is definitely wrong.¹ Similarly, in CSM 59:16:

e fre•mo•s' e de bon prez

it is important to keep the syllables mo and se, so that the s is syllable-initial and intervocalic, and therefore voiced, resulting in [fre.mo.ze]. If the syllables were broken as mos and e, this would imply [fre.mos.e], which is wrong.

[1] Note that this is the reason why written forms like quena and nono exist at all: although they are composed from quen + a and non + o grammatically, joining them together puts n at the start of a syllable, preserving the pronunciation [n] as a case of external sandhi.

Sorry about that. This is a known bug with "hinting" in the core Ubuntu fonts, which makes tildes appear flat at some point sizes. There's not much I can do about that myself, alas, but rest assured that actual macrons aren't used on this site, so there is no scope for confusion.