Apologies if this has been discussed before. My Geiriadur Mawr flops open at a page in the H’s and at the top I see the word hollalluowgrwydd, a reasonably long word which means omnipotence. It set me pondering what the longest Welsh word is, though of course on line searches just lead me to THE place name. So far I have found gwrthgyferbynnedd (oppositeness). What is the longest conversation word? I’d love to find one I could reasonably drop into conversation, though I imagine there may be some seldom long words, after all when did any one last use the English word antidisestablishmentarianism?
Here are a couple that I do use on occasion, but they are unlikely to be record breakers
cyfarwyddiadau - instructions
argymhelliadau - recommendations
I just did a quick-and-dirty search of a copy of the Termiadur’s wordfile that I happen to have lying around (as you do), and the longest non-hyphenated Welsh word in it appears to be “rhyngdrawsnewidioldeb” (22 letters, 4 more than gwrthgyferbynnedd) - which apparently means “interconvertibility”.
thank you, looking good, any advance on that?
On thing I’ve always wondered about that word: It’s a noun, right? So it should be possible to pluralise it, shouldn’t it? So you could make an even longer word just by sticking an s on the end.
Surely I can’t be the only person ever to have thought of that. Or do plurals not count?
19 letters in “RHyNGdrawsnewidioldeb” surely? 4 more than the 15 in gwrTHgyferbynneDD?
I would now put a smiley face and say “I shall now take off my pedant’s hat”, but it seems to be on too tight…
Dw i’ n meddwl: E…L…A…S…T…I…G
Ah! Pedantry! Digraphs, surely, Owain, those capitilisations, not letters…let me pull that hat a bit…
I have a feeling we may be fighting over this hat…
“Letter” and “digraph” are not necessarily mutually exclusive terms!
In Welsh, as in some other languages, digraphs can be regarded as one letter formed of two symbols which “coincidentally” resemble other letters.
Actually, saying that, I think that the definition of digraph may be a little fluid, but in Welsh, “t” is one letter of the alphabet, “h” is another, and “th” is another letter, a digraph, forming the fifteenth letter of the alphabet.
The letters have not physically “fused” in the same way that “uu” fused to “w”, but they are still regarded as one letter rather than two.
(Pulling brim of hat further down brow whilst clinging firmly to it with both hands…)
[actually, there was a tendency in mediaeval written Welsh for, eg, “ll” to have a connecting mark, but printing seems to have put a stop to that!]
*sigh* OK then - 22 characters, howzat
But this often comes up in length of Welsh words. How many letters in “Llanfair.p.g” and all that, which is different in Welsh!
It is an intrinsic part of the Welsh alphabet and the Welsh language, after all.
Myself, I would go for the length of a word being the number of Welsh letters, as in crosswords and- well, everywhere else!
But whatever floats your boat!
I guess. The whole digraphs-as-letters thing makes perfect sense in most situations, especially when you’re concerned with the actual values of the letters, but longest-word contests are basically just a frivolous “ooh, look at the big long word” type of thing anyway, in which case “which one looks biggest” could be said to be more important…
(Yes, I am just trying to justify this after the fact… ;-))
In the other news, the digraphs thing plays merry hell with collation. It’s impossible for a computer to correctly sort a list of Welsh words just by looking at them - you need a list of exceptions, such as “arholiad doesn’t actually contain a Rh”. Grrr.
Yup. Though I do have a sneaky feeling of pleasure in the idea of someone working in a dusty oak lined library with bits of paper filed all over the place winning out over a computer
But aye, that sort of thing is awkward with computers.
Mind you, just on the “rh” “r-h” thing, doesn’t the “rh” only occurs at the beginning if a word, where it will never be “r-h”?
If that is true, that might be easy to work into a programme- but other digraphs could be more awkward!
Let’s bore everyone by looking at the other digraphs in that fashion, as to whether a computer could cope.
“Ch” - well, don’t think you ever get “c”+“h”, do you? So that shouldn’t be a problem. Just assume digraph.
“Dd” - I when “d”+“d” occurs, it is hyphenated. No problem.
“Ff”- again, can’t think where “f”+“f” would occur, (could be wrong!) so you can just assume digraph.
“Ng”- now this one is a bugger. “N”+“g” often occurs, and there is no way to tell the difference, as it is not hyphenated.
“Ll” - again, “ail-law”, second hand, hyphenated, shouldn’t be a problem.
“Th”- can’t think of anywhere that “t”+“h” would occur, but that means nothing. If it doesn’t, no problem.
So it seems that it is only the fact that “n”+“g” is not hyphenated in spelling that would cause problems with writing a computer programme?
I know nothing about computers besides programming in BASIC, so that could be absolute rubbish!
But it seems it should be a surmountable problem, except for “ng”? Would cut the “exceptions” down considerably, if so.
[Actually, “n”+“g” should of course be “ng”+“g”, but that doesn’t help matters!]
As is often the case, it is all in the definitions, when it comes to pedantry. I think that Ifan was using ‘letter’ to mean a single character. Digraph and letter can overlap, as you mention. For instance, in Dutch, the digraph ‘ij’ is two separate letters for sorting purposes, but one letter (and is a ligature - fusion as you say) when written. Maybe if Welsh were to use ligatures with dd, th, ch, ll, ngh, ng, etc, we would not have this confusion. Could be a nice typography design exercise.
@louis Oh, absolutely! Definitions of such things are however you want to cut them, but the conventional answer to “how many letters are there in “rhywbeth”” in Welsh, counts the digraphs as one… Erm… Thing. Just Welsh convention, as you say!
I have little interest in defining "digraphs " or “letters” for other people, just nice to see Welsh conventions in these things!
Though not compulsory, of course!
And Welsh did have quite a selection of “one symbol” letters for its digraphs before printing was introduced.
The ligature in “ll”, the use if a form of “d” for “dd”, a single symbol for “ch”, even different firms for “w” as a vowel and “w” as a consonant, which would be a very useful thing to have nowadays! (Though the last was by no means used consistently, alas.)
With printing, those forms disappeared.
It would be good, as you say, to see them reintroduced at least in some forms of calligraphy and writing!
Why is there a banana in front of your message?
And just to demonstrate that my above posts are incorrect, I’ve just thought of the word “anrheg”, presumably “rh” letter in the middle of a word.
[but I still think you could probably cut the exceptions down to a manageable level! Probably. Possibly. Maybe.]
I’m not complaining about the banana, it’s a merry looking thing. Just wondering, that’s all.
Banana? That is Discourse’s rendition of a half moon. It seemed like a good idea at the time
Excellent deductive effort, Owain! You raise a good point, and inspired me to investigate further… (y’know, since we’re already boring everyone, we might as well carry on ;-))
I happen to have another wordfile here, extracted from Mark Nodine’s dictionary, who it turns out was generous enough to separate non-digraphed letters with a dot. So, based on that, I can confirm that (according to this dictionary at least), this only occurs with R-H and N-G, as you correctly predicted! And that there are 156 examples of N-G not being an Ng, and 61 examples of R-H not being a Rh.
Conspicious examples include: parhau, sarhad, torheulo, Aberhonddu; blaengar, dangos, pengaled, teyrngar, uniongyrchol, Bangor.
As for an actual Rh occurring in the middle of a word, there are 54, including: anrhydedd, olrhain, penrhyn and of course unrhyw! Though to be fair, those are all compounds and/or prefixed words. Now that I think of it, though, it is a bit strange that those Rhs didn’t undergo internal soft mutation…
I can’t not point out that antidisestablishmentarianisms would be a range of (probably rather similar) views held by…