Hello,

What an exciting prospect!

I am currently working with Feis Tighe Chonáín myself:

https://archive.org/stream/feistighecho ... 9/mode/2up

Your probably aware of these digitised resources, in case your not:

http://celt.ucc.ie/index.html (the digitisation of Old, Middle and Early Modern Irish edited texts):

Cath Mag Tuired: http://www.ucc.ie/celt/published/G300010/ (they have the English translation of some texts as well, if they have been edited and translated).

Feis Tighe Chonáin: http://www.ucc.ie/celt/published/G303010/

and


Irish Sagas online http://iso.ucc.ie/ (this could be very useful for your own work?!)

Where you'll find Old Irish sagas transcribed, and translated into Modern Irish and English:

Lists of sagas done so far: http://iso.ucc.ie/Irish-sagas-list.html

Aislinge Óenguso: http://iso.ucc.ie/Aislinge-oenguso/Aisl ... -text.html

Just a few questions about the description of the project: http://irishstudies.nd.edu/news/48812-t ... h-project/ :

"While there are many resources available for learning Modern and Old Irish, there are no comparable materials for learning Early Modern Irish, the written form of the language used from roughly 1200-1650. There is no comprehensive grammar, no guide to translation and interpretation, and no dictionary. Consequently, nearly 500 years of Irish writing remains grossly underused by scholars as the difficulty of acquiring the language limits access to a small group of specialists. The “Reading Early Modern Irish” project addresses this scholarly gap by offering the first systematic introductory apparatus for learning to read, transcribe and translate Early Modern Irish. ."

Are you focusing on Poetry (i.e. Classical Irish as well)?

There are grammatical tracts for Classical Irish, written by the master poets themselves. For instance, https://books.dias.ie/index.php?main_pa ... cts_id=379 and http://www.vanhamel.nl/codecs/Irish_grammatical_tracts

As regards dictionaries to use, there is the DIL (and eDIL) and Dineen's dictionary; the only problem is, because of the huge variation in spelling in this period, as with any period, you have to have an idea of how this spelling fluctuates to look things up. (e.g. often dropping lenition in medial -dh- and -gh- when using the DIL and dropping medial -dh- and -gh- in general when using Dineen, as -dh- and -gh- are often added despite there being no historical necessity to do so).

As regards tackling Early Modern Irish Prose, its such a mind field of different variables;

(a) Clear vernacular dialectal features emerge in writing, Early Modern Irish is very similar to Modern Irish itself.

(b) Scribes often attempt to use archaic features, such as using infixed pronouns, with varying degrees of success.

(c) Hypercorrection, a facet of archaising, such as attempting to supply or remove inorganic f, attempting to reverse univerbation and restore old compound nouns with varying degrees of success etc...

This leaves a very muddled language, with language varying from author to author depending on register and style implemented, i.e. whether or not he wrote as he spoke (then you are left with a language very similar to Modern Irish), or if he tried to bring in archaic features and ended up not getting it right and thus your left with a language that is neither Old, Middle or Modern Irish, e.g. Marco Polo.

Dictionary wise, you just use DIL and Dineen's dictionary again, you just have to be aware of how spelling variation works.

Have you thought of a pedagogical approach?

It seems to me - please tell me if I'm wrong and you have a different approach (I'm really curious!) - in order to navigate Early Irish successfully, given the wide variation in the language:

You need a firm foothold in Modern Irish, to get over hurdle (a) above,

but you also need a good basis in Old and Middle Irish to get around variables (b) and (c); as how else are you going to be able to identify attempts at archaising and hypercorrection? Especially, when the author is incorrect in his implementation of these features?

Cian

_________________

Please wait for corrections/ more input from other forum members before acting on advice

Thanks for the questions. Just popping in for a few minutes and I'll try to answer them all eventually.



Yes, absolutely. About half of our sample texts will probably be poetry (mostly Classical/bardic). The question of Classical vs. non-Classical was brought up at our meeting with potential advisors. So, basically, the main feature of the site will be a series of sample texts -- brief excerpts from larger works -- that have each word grammatically explained. In addition, multi-word phrases will also be parsed/explained. Will say more about the approach to explanations in a minute, but indeed, we did decide that texts should be marked as Classical or not Classical. Learners with just Modern Irish will have trouble with either register, but they would need to know that this particular text is Classical and this other one is not.



Yes, indeed. Eoin Mac Cárthaigh, who edited one of them, is working with us and has been a tremendous help so far. You're absolutely right that the grammatical tracts are available as resources (one of the main resources for scholars as well), but they're not of much use to the learner who isn't already either a budding scholar of Irish or an extremely talented and motivated learner. Clearly, some of the folks here would fall into that category, but the number of people who would find the Grammatical Tracts to be an efficient route for learning Classical Irish is still relatively small.

The user base/audience we have in mind includes, for example, historians and literary scholars who have decent modern Irish, but might wish to be able to use Early Modern/Classical sources in their work. In other words, we want to make Early Modern sources more accessible to people who wish to read them accurately and perhaps translate them as well. There is no pretension here of teaching people to be experts in the language. Our hope is that some users of the site will go on to become scholars/experts in Classical or Early Modern Irish, but the primary goal of the project is to give people the basic reading and translation skills they need to engage with the material.

Yep. Preaching to the choir here. Dil/eDil is great, but just overwhelming for most learners/non-linguists. And while Dinneen is useful, there are obvious limits, such as the ones you mention. Just in terms of the dictionary headword entries, the plan is to link to/cross-reference with alternate spellings of what are ostensibly the same word. I'm not the web developer, but the design of the search function will have to take alternate spellings into account in a way that is useful, precisely along the lines you mention here. Not sure how that will look on the backend, but definitely we want variant spellings to be returned when users do a basic search within the glossary/dictionary.

Great points. Many thanks indeed.

Mind field or mind field? Yes, both!

a) The emergence of vernacular features is one of the arguments we've made in funding applications for why this material is so important. And yes, it's very similar to Modern Irish, but it's also different enough that many fluent Irish speakers, scholars or not, have trouble with it. That isn't to say that one can't get a fairly good impressionistic understanding of Early Modern Irish prose, but good, solid reading proficiency and comprehension is a different matter. Historians, for instance, who have Modern Irish, may be able to read Keating and understand 90% of what they read, but one inevitably encounters constructions or mutations or... you name it, that are perplexing because they aren't typically found in Modern Irish.

b) We've talked about archaisms, infixed pronouns and the like. Because the main pedagogical feature of the site will be the sample texts, such archaisms can be pointed out, explained, etc. whenever and wherever they appear. Theoretically, one could give the user a set of rules to learn which could then be applied to actual texts, but not many users learn in that way. And, as I think you're suggesting, how does one devise rules or abstract explanations for dealing with both "proper" grammatical usage as well as deliberately archaic (i.e. incorrect) grammatical usage?

One way around that is to rely on sample texts for the most part, where discrete instances of actual usage are explained in context. Just as building the website will be an ongoing, iterative process, users will build their reading skills through an ongoing, iterative process. Fortunately, we don't have to worry about writing a comprehensive "code" that the user will then apply to individual texts. It should actually work in the opposite direction, where users read different samples texts, along with their grammatical explanations, and then slowly begin constructing an understanding of the "code" over time.

c) Again, individual instances of this will be highlighted and explained in context through the grammatical explanations that go with the sample texts.

So, the glossary/dictionary will actually be constructed by digitizing and collating entries from glossaries appended to various printed editions of Early Modern texts. Take any Irish word you like, and it is likely to appear in the glossary for several different modern printed books, whether it's a miscellany of bardic poems or the poems of a single poet or a prose work in Irish, etc. The idea is to take each glossary entry for a particular word and collate them. So when you look up a word in the glossary, you will be given each glossary entry (drawn from various books/editions) for that particular word. There are two advantages to doing it this way: 1) it's fast and efficient (no one has to write a dictionary entry for each word) and 2) the definitions found in glossaries are typically shorter and more learner-friendly than what one would find in eDil, for instance.

A conventional, full-fledged dictionary specifically for Early Modern Irish would be great, but that's a different project altogether.

It might make more sense once we have the site up and running with our first one or two sample texts, but the sample texts will have a series of hoverboxes, both for each word and for each phrase/semantic chunk (so, two levels of explanatory hoverboxes). The hoverboxes will give a very straightforward grammatical explanation of the word/phrase, but especially where complex issues are involved there will be a link from the explanatory hoverbox to the relevant entry within the stand-alone grammar. For instance, if there's an infixed pronoun, the hover box would highlight and briefly explain why it's there, but there would be a link from the hoverbox to the entry within the grammar that talks more in depth about infixed pronouns and how they are used or mis-used in EM Irish.

Again, rather than a genuine chicken and egg problem, where one has to learn Modern Irish "code" as well as Old and Middle Irish "code" before being able to read/decode Early Modern Irish and understand the relationship between them, we have an iterative process in mind where users gradually build their skills and knowledge over time by working through several sample texts. To mix metaphors even further, the user doesn't have to have all the pieces of the puzzle from the get-go in order to begin understanding what the finished puzzle will look like.

But your question raises another question that was raised in Dublin a couple of weeks ago -- how do we make sure that the sample texts contain the kinds of anomalies that users are likely to encounter in their reading beyond the website? Just thinking statistically, a couple dozen sample texts are likely to contain a high percentage of the most commonly occurring grammatical anomalies, archaisms, etc. That is, if they're fairly common, then there's a better chance that they will be found in the sample texts. By contrast, if the anomalies in question are rare, then it matters less for most users in most situations. Over time, as more sample texts are added to the site, a greater and greater percentage of the archaisms and other oddities will be covered by the sample texts. Being 100% comprehensive in terms of grammar oddities would be great, but it's not necessary for achieving the main goal of the site, which is to make Early Modern texts more accessible to more readers. And again, over time, more and more of the archaisms and obscurities can be added to the site, whether in sample texts or within the grammar.

Wes, how much in the way of texts is yet to be digitized and has there been much corpus linguistic analysis used on such material (to find patterns etc)

_________________
__̴ı̴̴̡̡̡ ̡͌l̡̡̡ ̡͌l̡*̡̡ ̴̡ı̴̴̡ ̡̡͡|̲̲̲͡͡͡ ̲▫̲͡ ̲̲̲͡͡π̲̲͡͡ ̲̲͡▫̲̲͡͡ ̲|̡̡̡ ̡ ̴̡ı̴̡̡ ̡͌l̡̡̡̡.___

James, are you talking about the couple dozen (or so) Sample Texts (with all the attendant grammatical explanations for both words and phrases) or the glossaries from printed editions?

We have two Sample Texts very near to completion, but the digitization of the glossaries has only just begun. We were really holding off on the glossaries until we had the proper permissions, but it looks good on that front. Unfortunately, OCR just isn't an option, especially with the sean-chló. I know that some folks (Kevin Scannell maybe?) have been working on OCR for Irish type, but even with Roman type, you'd need to go through and make corrections by hand. Basically, after some trial and error and some pretty good back-of-the-envelope calculations, we decided it would be most efficient to simply have folks transcribe (i.e., type) the glossaries by hand. We've some great graduate students at UConn working on that and it's gone well so far.

As far as corpus linguistic analysis, that's a bit above my pay grade, but I'm not sure our project would be ideal for that kind of thing. The sample texts, for instance, are very brief and there will be a relative few of them. I would think the digitization efforts associated with CELT would be more suitable for that, but I'm interested to hear what your thinking on this is as my understanding of corpus linguistics is fairly vague and abstract I have to admit.

On the second point, I meant yeah, digitisation as I wondered if once it is all up, not for the site (which would be looked at by eye and annotated and packaged), but for later, would computer analysis be brought to bear on the corpus of text, say to make it easier to check word forms that were regular (grammatical) or unusual (quasi-archaic, such as in=fixed pronouns).

I didn't make that clear, sorry

[I see you answered the question :]


I'm not an expert on the field, but I would guess it might be tricky due to the liberal attitude to spelling that our forefather in English and Irish seem to have had (Aiden Doyle in A History of the Irish Langauge makes mention of one word spellable over a thousand ways in the pre-reform orthography, and Shakespeare's works are said to be replete with variance), so deciding on a base head-word would be tricky. I guess words would have to be treated as more abstract entities than tokens, like a cloud of potential forms (beirbhiughadh, beirbhiú, beiriú, beiriughadh, plus the ponc forms, lenition & eclipsis etc)

_________________
__̴ı̴̴̡̡̡ ̡͌l̡̡̡ ̡͌l̡*̡̡ ̴̡ı̴̴̡ ̡̡͡|̲̲̲͡͡͡ ̲▫̲͡ ̲̲̲͡͡π̲̲͡͡ ̲̲͡▫̲̲͡͡ ̲|̡̡̡ ̡ ̴̡ı̴̡̡ ̡͌l̡̡̡̡.___