119: Irregularity is all over the place

Meat clarinet merch and zines

0:00Welcome to Lingthusiasm, a podcast that's enthusiastic about linguistics. I'm Gretchen McCulloch. And I'm Lauren Gawne. Today we're getting enthusiastic about irregularity and the consistent inconsistency of language. But first, we have new merch.

0:32Okay, backstory. We've used the metaphor of a musical instrument several times when talking about how the human vocal tract is organised and how you can change the shape of your mouth as the air is on the way out, which changes what the air sounds like, which is true for both the mouth and a musical instrument. We've used a couple of different musical instruments as metaphors – the mate clarinet, the mate bassoon, the mate bagpipe. But I just can't get that original imagery of the meat clarinet out of my head. Maybe

1:03it's because I played clarinet in high school. Who knows? So Lauren has very generously allowed me to persuade her that what we really need in our lives is some art of the human vocal tract and a clarinet merged together in meat colours. To be clear, you need this in your life. This art is deeply cursed and I will not be buying any. This art speaks to my soul. Our artist Lucy Maddox did an amazing job,

1:36knocked it out of the park as usual. And Lingthusiasm patrons can even see behind the scenes some sketches of the several horrifying meat textures that she offered us as we were coming up with this design. Truly one of the most terrifying design briefs I have ever had to write. Go have a look at the meat clarinet available on all Lingthusiasm social media and tell us if you love or hate it. As people have pointed out, this is not actually an anatomically accurate drawing because

2:06the mouth at the end of the meat clarinet would have to be your mouth. That was even more disturbing if we tried to line all that up. Just let me have my body horror in peace, okay? The meat clarinet is available on stickers, mugs, t-shirts, other horrifying objects. You can follow the merch links on our website or social media to see it for yourself and decide if you are hashtag Team Lauren or hashtag Team Gretchen on this front. Something I am really excited about is this

2:38is the first time we have launched merch with its very own zine. A zine is like a miniature DIY magazine, hence the name, which you can reproduce yourself on a photocopier. We've been experimenting with making zines about linguistics over the past couple months so that we have something to give out to people at conferences and other types of events. We're making all the Lingthusiasm zines available as PDFs on the Lingthusiasm website so you can print them out yourself if you want to spread the joy of linguistics. We've heard from a lot of listeners

3:09over the years that Lingthusiasm helped you get into linguistics and now you're trying to share that with your friends, family, classmates, but sometimes it can be a pretty big ask to get someone to listen to a whole half-hour episode so what if you could just hand them a fun little mini zine with illustrations as an easier starting point? We've shared a couple of zines on Patreon already and Gretchen's been sharing production updates through Patreon's Quips feature if you want to be in the loop on this new little project. You can see future zines and you can give feedback before they're live for everyone. Our first three zines are one from me about Boob and Kiki, a zine from Lauren about

3:45gestures – surprise, surprise – and an extra teeny tiny zine about making words for small things from our technical editor, Leah Velleman. Thanks to everyone who's been following along for your enthusiasm about the zines. We love seeing photos of where they show up in your lives. You can print off a copy or two to put on your bookshelf, your coffee table, your bulletin board, wherever people might see it. A few people have already told us that they are leaving them in little free libraries or putting them on zine swap racks, which I just think is so cute.

4:15Cute. I've been using the photocopier at work to print out copies to share with colleagues and students for educational purposes. Of course. If you get inspired to make any linguistic zines or you know of other existing linguistic zine projects, let us know. You can read the new Meet Clarinet zine and all Lingthusiasm zines so far at lingthusiasm.com. Our most recent bonus episode was all about poetry and how the structure of language shapes the structure of the style of poems. To listen to this and over 100 bonus episodes, go to patreon.com.

Pattern recognition pop quiz

4:49Slash Lingthusiasm. Language loves a pattern. Gretchen, we're going to do a pattern recognition pop quiz. Are you ready? I love a pattern too. Dog and dogs. Cat and cats. Chicken and… Chickens. Great work. Okay, now. As you might have guessed from making a podcast in English, I am in fact an English

5:24speaker. Okay, now. Child and… Children. Correct. Not childs. Not childs. Sheep and… Sheep. Yes, unless you're talking about many breeds of sheeps, and then I guess that's possibly okay on analogy with fishes. Person and… People. Yes, although I do notice when there's police reports and they talk about persons unknown. Right, or sometimes peoples of the world to refer to ethnic groups or cultures.

5:57Oh, yeah. Pluralising the plural is extra nifty. Thank you. Great English pattern recognition, including when the pattern is not regular. Let's take it up a step. Let's see how your pattern recognition goes for Yolmo, a language – as far as I'm aware, you do not speak. Everything that I know about Yolmo, including that it is a language of Nepal that you have done fieldwork on, I have learned from you on this podcast. Great. I'm very glad to hear that you haven't been withholding your Yolmo speaker status from me. I would be quite upset. Absolutely not. What a secret.

6:30I'm gonna give you one verb and the different ways that you can make that verb, and then I'll give you another one to see how your pattern recognition goes. Okay. See if you can spot where the pattern doesn't stay patterned. The verb eat, we have sage, saterang, sarsin. Okay. These are eat in what kinds of eat? That's eat in the present, eating, and ate, which is not regular in English. Okay. But this is regular in Yolmo.

7:00Yes. And now we have the verb go. So, doge, doterang, kalsin. So, sage and doge both seem to pattern with each other. They both end in ge. I eat, I go. I eat, I go. Saterang and doterang both have this terang at the end. Eating, going. This looks pretty regular to me. And then we have sarsin and kalsin. And this do that I was seeing for go all over the place is now kal, which looks completely different to

7:32me. Yeah. A little bit like when I translate it into English. So, we've got I go, I am going, I went. Also very regular. Yes, that's true. Yes. And so, you have a completely different – it's not like it's a little bit the same, like eat and ate. It's a completely different form popping up in the past here. And it's just the case that what is this kind of irregular in Yolmo for this word is also irregular in English in this way. That's such a neat coincidence that these two languages are irregular with respect to

8:04themselves for these forms, but then have this macro-regularity with respect to each other that I assume is coincidental. But also not surprising. Yeah. Because irregularity in this way tends to pop up on frequently used forms. And if there's one thing humans like to do, it's tell each other if they're going places, who's going where, what's up, where someone went. And so, it's not surprising. But also, I love your very classic linguist enthusiasm for finding irregular forms in a language.

8:37It's so fun. It's like going to someone's house and they've got a weird little nook or a little stained glass window somewhere or just some sort of interesting, quirky piece of decor or architecture. I just get so pleased with it. Indeed. And I feel like, especially for verbs, this kind of irregularity is so common, it's a bit of a running joke among linguists. So, there is a meme that circulates. Do you mean the classic meme, tense, moody, irregular? You must be a verb.

9:09Indeed. Yes, I do. But as we'll discover in this episode, it's not just verbs. In fact, irregularity pops up all over language. Even though having things in neat, ordered, replicable patterns is really like languages, modus operandi, irregularity also pops up all over the place as well. I feel like my opinions about irregularity have not always been quite this positive. When I first started trying to learn languages and

9:41being like, oh, I've got to learn all of these forms or all of these exceptions, I was a lot more like, why? Why am I doing this to me? This is a horrible prank perpetrated by this language on me. Personally. But now that I have the more linguist brain galaxy view of the patterns between the irregular forms and what is irregular across languages, it just seems so much cooler. And just knowing there's no perfectly regular natural language, that this kind of feature arises just because of how language is transmitted from one generation of speakers to the next.

10:15And that when people have tried to create one, I think Esperanto being the primary example of this, which was created in an attempt to be a very regular language that would be easy to learn, it turns out that as people actually learn and use it, it evolved irregularities. And this seems to be something that evolves naturally even when we try to regularise it out because it seems like it ought to be easier. So, in a fascinating twist of fate, the tendency to become irregular is in fact one of the most deeply regular things you can say about language. So, let's take a tour of some of the ways that languages can be irregular.

Sound system irregularities

10:50Sure. So, this can start in the sound system and the sounds or the handshapes that people use to create a language. And the sound system is a really good place of illustrating how irregularity has this sort of push-and-pull effect where sometimes the language gets a little bit bumpier and then sometimes the irregularity gets smoothed out again and it seems to go back and forth. One of my favourite examples of this is a word you actually introduced me to way back when, which is the English word sneeze.

11:21One of my favourites, which we now say is sneeze because phne is just such a non-common – it was so irregular that we just freaked out about it and couldn't phnees. We couldn't phnandel it.

11:37Sometimes I think about the old history of English where we actually instead introduced a whole bunch of other words beginning with phne and how phneaking phneul that would be. But alas, this is one case where an irregular form was so irregular and so marginal that we lost it. But sometimes we can produce some irregularity. So, English has click sounds as a tsk or a get someone's attention type sound. We don't make words with those. But it's possible

12:08that they could be moved into the sound system in some alternate universe. Right. I've seen people sometimes call them verbal gestures that are often accompanying a gesture or not necessarily having the same status as the grammatical part of the system. Often include, in various languages, sounds that aren't part of the regular words in that language. Even though there are totally languages – Zulu and Gosa being among them – that use clicks as part of their regular words, just as you might use a P or something in English.

12:40The way sounds are spread out in a language can be irregular, although in English it happens in very regular ways. The ng at the end of words like hang is only ever at the end of words in English, whereas it can be at the beginning of words in other languages. And, you know, if we had enough contact between English and Vietnamese where ng is a totally reasonable beginning of a word, we could end up borrowing enough words from Vietnamese that this would become possible the beginning of words in English. Okay. We need to add like 100 words from Vietnamese, 100 Tibetan words, and maybe like 100 Australian

13:13Aboriginal language words and like really cement this and smooth out that irregularity. That sounds like a fun project to me. In fact, that is sort of what happened with English and the V sound. See, there's precedent. My plan could work. English used to only have V between vowels, like in leaf and leaves. So English couldn't say vowel or verb? Nope. Like in leaf and leaves, which used to be pronounced leaves. So there was a vowel there,

13:47I promise. Hence the spelling. And in Old English, this is what was the case. And then after the Norman conquest, there were enough borrowings from French that did begin with V, like words like vowel, but also words like vine, which contrasted with English fine and introduced V in a lot of other positions. And now we think of V as a totally separate sound in English, except for these like quote-unquote irregular plurals, like leaf leaves, which used to actually be regular sound change. Hmm. Right.

14:17So sometimes regularising one area ends up irregularising another area. I feel like when it comes to irregularity, we do just have to do a shout out to the English writing system, which feels irregular because English is such a borrower of a language and is so interested in keeping something that looks more like the original language than necessarily reflecting the way English speakers would pronounce something. English really prioritises etymology in our spelling over pronunciation.

14:51Hmm. And this is one kind of irregularity at the expense of another, really, because it makes the connections between the words from the other languages easier to spot. It makes the connections between like photograph and photographer, which have totally different vowels but are clearly related to each other in terms of roots. It makes those connections easier to spot, but then it makes the actual pronunciation of words harder to spot. But sometimes we can't cope with the irregularity. Yeah. So, it was very heartening for me as a child who really struggled to spell a word like

15:26could. It's got a silent L in it. Yeah, where does that come from? It's got a silent L, but that L doesn't even exist in there historically. But it came in because of analogy with would and should, where the L is still silent, but it's there etymologically because would and should are the past forms of will and shall. Hmm. And then could is the past form of can, which as you know, doesn't have an L in it. Not cow. Not cow. But the L gets added on analogy once the other ones have become silent. So, even though it's

16:00not an etymological L, it's still an analogical L and has some function that makes sense at a different level. So, regularising this weirdly irregular quirk of historical reference. Oof. There are so many other fun kinds of irregularity that I don't want to get too caught up in spelling because, fortunately, we have a whole episode about spelling and why it's complicated. So, we can send you to that instead. I feel like there's also a bunch of lexical irregularity and irregularity around words.

Lexical irregularity and euphemisms

16:35I love the phenomenon of unpaired adjectives in English where you have something like kemped or couth or ept or shoveled. I feel like some of these are almost making a comeback with how often my friends love using the unpaired positive form of unkempt, dishevelled, unseemly. Uncouth, inept. And some of them are preserving old sound changes. So, unkempt is the opposite of uncombed and inept is the opposite of unapt. So, there are some sound changes that have made

17:10it less clear what the positive form is. There's also something more generally irregular and uneven around the way we have so many more words for negative things in English. We love constantly churning through and deciding that something is not polite enough or we have lots of different ways to say that something like smells bad, which creates this irregular balance in how many positive adjectives or words we have and how many negative ones we have for particular phenomena.

17:41This is related to a phenomenon known as the euphemism cycle, which was named by a linguist named Sharon Henderson-Taylor in 1974. So, the euphemism cycle has been going on for a while, where as something is associated with a negative word, it feels like it takes on those negative properties in the minds of its audience. Then people start using another word to have fewer of those negative connotations. Because the original concept still has those negative connotations, eventually, that word too becomes tainted by the negative connotations of being dirty or

18:15otherwise stigmatized. Then we have to invent another word that becomes a little bit safer as a way of returning to that concept until it too joins that cycle. You see how different forms of irregularity are created or repressed or expanded because of different mechanisms. This is one really productive one for words and why we have more words for some things than others. Can we talk about my favourite kind of irregularity?

Grammatical paradigms and verbs

18:42Sure. I think it's time. You've been very patient. This is irregularity of grammatical forms, also sometimes called paradigmatic irregularity.

18:57We've talked about this a little bit at the beginning of the episode. The sort of cats, dogs, children, goes, going, went type of irregularity. There is just so much of it in different languages. Indeed. With these grammatical forms, you often have a neat little grid of options. You've done cat and cats, dog and dogs. For verbs, you've got walk, walks, walking, walked. This neat little grid of options is called a paradigm. Some words every time will just fit neatly

19:30into the paradigm. Play, plays, playing, played. When you have fancier words, ironically, the more fancy and less common a word is, the more likely it's going to neatly fit in that paradigm. Compose, composes, composed, composing. If we coin a new word, it's very likely to be found in the same regular paradigm. Something like, you know, smize, smizes, smizing, and smized, you know. Don't use smize at me.

19:59That's, you know, more recent words than more likely to be in this very regular paradigm. You do have some sort of relatively rare archaic words that sometimes are still irregular but have been around in the language for a long time. I'm thinking of something like smite, smote, smitten. Cute. I always forget that smitten. I always think of it as like its own thing. I always forget it's the past tense of smite. Yeah, it's like literally lovestruck. I love it. Hmm. Which makes sense if you think about it as belonging in its own little sub-paradigm grid with write, wrote, written. Exactly.

20:31But it's not the bog standard one. In fact, like frequent, the thing that's really hard about this is you actually do want to reach for more fancy verbs or words when you're going through this paradigm because the more common something is, like, eat, ate is a regular, go, went is a regular. In fact, there's a study that looked at the 20 most frequently reported verbs for parents of young children who are learning English where fully 50% of those 20 verbs, fully half, are

21:02irregular in some way in the paradigm. That's not an accident because part of the reason they still are irregular is because we use them so much that we remember that they're irregular. Some of the ones, you know, things like smitten, which gets preserved primarily in this particular, like, adjective context, I wouldn't be surprised to say you hear someone use smited as the past tense of smite because we've kind of forgotten, or at least some people have kind of forgotten, that it had other irregular forms. Delighted to report an insufficiency of people being smited.

21:36Whereas for writing, we talk about this all the time, and so people are unlikely to forget that wrote or written are the irregular past tense forms. Yeah. English is pretty weak on the paradigm game, I have to say. When I was learning French in school, we had a whole book called a Becherelle, which is just like 250 pages in this little compact hardback volume that is about the art of conjugating and has all of the

22:07different verbs in all of their different forms, in the different tenses, you know, the subjunctive and the, you know, this one has a double L in it, so it has to be its own thing because sometimes the L is doubled and sometimes it isn't, and this directory where you could look up all the verbs, and we spent so much time with our Becherelles. What does Becherelle mean? Well, I just learned it at the time for the name of this particular kind of book in French. But I looked it up for this episode, and apparently Becherelle was actually a guy.

22:39Louis Nicolas Becherelle, born in 1802 in Paris, was a French lexicographer and grammarian. With help from his brother Henri, he wrote the first of these reference guides to French verb conjugations. Then he wrote a second one. It became this really popular version of a Becherelle, and now people often use a Becherelle as just the generic term for a, like, conjugation book in French, whether or not it's produced by Becherelle and his successors, but they're still

23:13being sold. Amazing. It's a bit like that period where a whole bunch of people just called their dictionary Merriam's dictionary after Merriam-Webster, where Merriam was a real person and Webster was a real person, but it just became a generic term. Yeah, like, a Webster was just a dictionary. And yeah, Becherelle was a guy. I remember – I still have my Becherelle. It's sitting on my desk. I'm looking at it right now from high school. I haven't cracked it open in a while because these days when I used to do the conjugation of verb, I can just look it up online. Spellcheck has a lot

23:43of them. But apparently, they're still being sold. I expect probably for the same use that I used mine for, which is that we were allowed to have a Becherelle for French tests in the exam room because you couldn't just let people look up anything on the internet, but you could look up the relatively limited conjugation table. That's probably why apparently millions of them have been sold in Canada. I would not be surprised. Whereas for an English learner textbook, you might find one page at the end that lists those 10 to 20 really common irregular verbs that people are going

24:13to encounter. Whereas in English, there's just so much less happening with what you can add into a verb and so much less irregularity in that that we don't need a full Becherelle. I've actually been volunteer teaching an English as a second language class, which is a fun activity for recent arrivals to Canada that I've been involved with lately. The textbook that we've been using has one page of here's the irregular verbs you need to know – eat, ate, eaten, go, went, gone. Amazing. There are a few others that aren't on this one page that are rarer verbs that they've

24:46decided the students don't necessarily need to know yet. But it's probably two pages. I remember trying to find a Becherelle equivalent for English and just being like, this doesn't exist. This can't exist. We don't need it. We don't have so much going on. If your language has a gender distinction on the verbs and that has irregularities, you're going to have to account for that. That just doesn't come up in English. A lot of the ways that English makes compound tenses are things like, I would have been able to go. When you have all of those words

25:23being stacked onto each other or I had been being perceived, the had been being part does make a compound meaning but isn't irregular because they're all individual words that are stuck together.

Pronouns, nouns, and fossil traces

25:36We've been giving verbs a really hard time. But to be clear, it's not just verbs that show irregularity. I mean, a big shout out to pronouns where the first person forms in our little paradigm are I, me, my, mine. But it does trace back to Proto-Indo-European. So it's been irregular for a very long time because that's a very long time in human language. You get ego, me, meus, things like that in Latin. You get this sort of parallel showing up in a lot

26:07of different Indo-European languages historically. Similarly, for other parts that aren't quite regular in modern-day languages, they trace back to something that was also irregular a long time ago. For nouns, if you take the word for hand in French and Spanish and Italian as well, I think, the word is la main or la mano, which is a feminine word that looks like, from the ending, like it should be a masculine word. Like en and o are both very masculine-looking endings.

26:42And that's because it comes from a Latin word, manus, which was already looking like it should have a masculine ending but actually being a feminine word. So it's preserving this aspect of the older language, even though it's not something that looks irregular from a one-day view if you're only looking at the present day. It's so great how irregularity becomes this fossil trace that lets you tell this bigger story of a language. When I was learning Italian very briefly, we got introduced to this unit of gender-changing

27:15body part nouns. Where there's a whole bunch of body part nouns in Italian that are masculine in the singular and feminine in the plural, like il braccio, which becomes le braccia, which means arm, arms, or il ditto, le ditta, which is finger, fingers. And I took one look at this and I was like, I bet these words were neuter in Latin because the neuter plural in Latin, like datum, data, looks like the feminine. And modern-day Italian doesn't have a neuter anymore,

27:50but because body parts were so frequent as words, they didn't quite get ironed out into looking regular. And I looked up the etymology, it turns out, yes, this is what happened. They didn't quite get ironed out into being regular because they were so frequent that the language could support this seeming irregularity that tells a story about where these words came from. And a lot of our sources of irregularity reflect little fossils of these older systems. So like, buy, bought, teach, taught is an older form of doing tense change in English that we've only kept

28:24in a handful of verbs. And the cool thing for me about learning German – this is really turning into Gretchen's language learning adventures – was that a bunch of the words that I'd always thought as these sort of weird irregular pockets in English had a similar irregularity going on in German because this vowel change alternation is a systematic Germanic thing. Sometimes sources of irregularity can be not because it stayed stable, like that stayed stable since something before English and German were separate languages. Sometimes it can be because things are coming into the language. Like, I think a lot about how people can get anxious about

28:59whether the plural of emoji is emoji or emojis. And it depends on whether you are treating it like other Japanese words that we've brought in, like sushi, which we keep in the singular and treat like a mass noun. So you don't say sushi in English, you say pieces of sushi. But with emoji, we're at this point where it's like, there are many emoji or there are many emojis, and they're kind of both up for grabs at the moment. When I was writing Because Internet, people were asking me, like,

29:30oh, you're writing a book about internet language. Like, should I pluralize emoji as emoji or emojis? And at the time, I chose to use emoji as plural for that book because that was what people who are more familiar with the emoji history in Japan were doing. And it sort of indicated that I was familiar with that history as well. But I honestly think that if I were writing it now, I would be more likely to say, look, the pendulum is swinging. More people are saying emojis now. Maybe I should go with where usage is currently heading rather than preserve this more historical approach. And that hasn't been

30:04that long since I wrote that book. But things change. We also sometimes make things more irregular to match up more closely with something that already is irregular, so on analogy with something that already exists. So, the older and more historic past tenses of dive and sneak are actually dived and sneaked. But you would have a hard time convincing modern-day English speakers of this because dove and

30:35snuck have become so common by analogy with similar English past tenses like teach and taught or thrive, throve, that dove and snuck feel like they should just be there because they're like other words in English. So, the answer to why things are irregular is that sometimes we like them. The existing irregular stuff has a gravitational pull. And while I don't think this is necessarily a productive

31:06source of irregularity, I do just need to give a big shout out to the most important category of irregularity, which is jocular irregularity, where people deliberately play around with the form of a word. I think this is so much fun. I think that the delightful absurdism in saying, you know, octopodes or octopodes, which really sounds like a hidden Greek playwright has one with eight arms.

31:37It's just – it's so fun and people like extending octopodes to other types of words. Yes. So, once you've got octopus to octopodes, you can have glamourpuss goes to – look at all those glamourpodes. Ah, amazing. Or there's a period in like computer nerd history when it was very popular to use boxen as the plural of box by analogy with ox oxen. Like, come on, that's great. That's so fun. Yeah, a very important and often underappreciated way of showing linguistic

32:10competence by playing around and messing around with the paradigms, but not necessarily something that stays productive. Sometimes I start doing something and it's for a bit and then five or ten years later I'm still doing it and maybe it's not a bit anymore. Yeah, like, I really think the jury's out with octopus and its various plurals. Check in again in 50 years and we'll let you know.

Suppletion in language families

32:33So, we've talked a little bit about go and went. It's come up as a couple of examples. And I thought I'd just flag that this has a very specific technical name in linguistics. I love all of the kinds of irregularity, but this kind is especially cool and has a very special place in my heart. Is this your favourite form of your favourite form of irregularity? Is that how far we've come? Oh, yeah. I think this is where we've come. So, when part of your paradigm just replaces – instead of changing the affix or anything,

33:09it just slaps a whole different word in there with a whole different form and often a whole different history – that is known as suppletion. So, when you have something like bring brought, you can still see that they're both clearly like bruh words. They've got a common history. It's just a different word change pattern. But go went or good better or bad worst really look like completely different words. It's only that we're so used to them as English speakers that we're like, yeah, these are just variants of the same word, like obviously. This is because at one point in history, they were genuinely completely different

33:42words that have smushed themselves together into different parts of the paradigm. Went is the old past tense form of wend, which like wend one's way, like move. So, it's still a word that has a similar semantic sense to go. In Yolmo, that kal suppletive past tense for do – so, do is the regular verb to go – crops up across all of Tibetan. It's only in this branch of the family where they've taken a word that meant like – to cross. It's like to cross a bridge or

34:17something. My theory is like if you're leaving one of these villages, you're leaving via a bridge because we're up in the hills. Yeah. Maybe especially if the leaving has happened in the past, it's more likely that it was like a bigger trip. I don't know. Who knows? But it's a – they crossed – somehow became the preferred form of they went, similar to English. Do you think wend one's way is related to wind? Yes, it is. I did check the etymology of wend before this. So, like wend and wind – you can still

34:49see all these like historical threads for wend, and we've just jammed it into one of our most common verbs as part of its paradigm. Amazing. Also, just a big shout out to copulas. This is the is, are, were, being massive cluster of suppletion in English. Copulas do tend to be a bit of a suppletion or even just irregularity magnet because they are such high-frequency words in languages that have copulas. We have a whole episode about copulas as well because there are so many different

35:23areas where irregulars show up. In many of the languages that I have studied, especially any of them that have been Indo-European languages, there's been irregulars in the copula form. There's been irregulars often also in the go form as well. The romance words for go like aller and je vais and irer, these are all French forms of the verb go, and they sound completely different from each other. There tend to be some things that are commonly found to be irregular. We've name-checked a whole bunch of them in this episode. In signed languages, negation is a huge magnet for irregularity.

35:59In Auslan, really basic verbs, the positive and negative forms will have something different about them, but it's not like a neat n't or not or no that's very neat as a paradigm in English. Can is like you draw your index finger and thumb away from your nose in a straightaway part, and then can't you rotate them against each other. Same articulators but a different movement. And that's not systematic across other negative forms?

36:32No, it's like want and don't want. Again, same hand shape, different direction and motion, like and don't like. All these things, the positive and the negative aren't just a very regular paradigm of the positive and the negative. They're all doing something slightly different. That's fascinating that you'd say that because in ASL, that's also true. Can is produced with the fists up, bent at the wrists, and can't is produced with the little fingers out, striking against each other. This is not at all systematic.

37:02Oh, so completely different. It's interesting you say that because there's a World Atlas of Linguistic Structures survey from Ulrike Zession, which is all about the prominence of irregularity in verbs in signed languages. There's like, of the 35 signed languages surveyed, literally one has no irregular negatives. I'm starting to think that this is not an irregular pattern at all. This is a highly regular pattern. 21 of the 35 have more than five verbs that show this

37:33kind of irregular pattern. This is just a massive locus of common irregularity in signed languages. There are a couple examples of negative irregularity in English. I'm thinking of something like will, won't, or shall, shant, which are sort of the two that I can think of where there's this irregularity with the L. But this is not the same kind of systematic common across signed language families, not even related signed languages. I guess there's also English weirdly doesn't have a contracted negative form with am, except

38:09in some varieties. We have isn't and wasn't and aren't and weren't, but we don't have amant. At least, my variety of English doesn't have amant. I hear that it's found in Irish. Shout out to Irish and those varieties of American English that are re-regularizing and keep amant in their paradigm. Big ups to you. I've also heard that ain't originally started as amant and then got extended to the rest of the paradigm and then denigrated. Sometimes in stigmatized forms of language, bits of irregularity

Child language acquisition

38:40can pop up. Irregularity has this real interesting effect when we see how especially children acquire irregular forms. There's such a common trajectory that we can map across all kinds of irregularity. At the start, a child is using words with no additional morphology. They'll just say shoe or child, whether it's one shoe or 20 shoes. There's no marking of that at all. Then at the next stage,

39:14they do regular and regular plurals correctly. They look like little language learning geniuses. They'll say shoes. They'll say children. You're just like, great. You are learning language. Great job, child. Yep. But then they appear to go backwards. They'll start saying child instead of children. And you're just like, oh, no, but you could say that properly before. But it appears at this stage where their correct usage drops is where they're actually learning shoes is the default form.

39:48They've learned this rule and they're applying it. They're just over-applying it now. I love how the sort of quote-unquote mistakes that people make can tell us something about the languishism that's going on in their heads. Yeah. It actually shows that they're really acquiring this as something productive. They're applying to shoes, childs, pantses. Just like, you've got to really, really apply it in all contexts. Then it's only at the end that they go back up to being able to competently say shoes and

40:20children. They relearn, oh, this one's actually a regular. I've got to have my special role for this. They go from being competent to, if you're plotting correct usage, they drop and then it comes back up. This is known as a U-shaped competence curve. Because it makes that shape of a U. I just think this is so nifty. I think it's really important. I don't want to be like, and then at this age they do this, and this age they do that. Even some of the earlier studies where they were plotting this out, it was just like, oh, this kid who's a few months past two can do the correct plural 98% of the time

40:55at the same age other kids in the study were doing it 30% or 13% of the time.

41:04It's not about hitting milestones at particular times. It's this process of learning each word individually and not having really a system, learning the system and over-applying it, and then returning to that irregularity. It's such a fun thing to observe if you have any kids in your life who are going through that stage. Yeah. Shout out to Carolyn Rowland, whose textbook not only summarizes, but she picked out all the really fun examples like pantsers. I really appreciate that level of textbook quality

41:34summary. I also think this is relevant from a language learning perspective because sometimes I feel like I'm jealous of kids who are learning their language what seems so smoothly, but they're actually going through this trajectory of trying to learn the irregular forms and over-regularizing and then coming back and learning them. I feel like it's helped me be more relaxed about sometimes I'm not going to get all of those forms in the paradigm correctly. People still understand children. They can still understand me as I'm gradually figuring out how to do that.

42:09I'm really glad this process has helped you make peace with your language learning experiences.

The linguistics meme evolution

42:14I think perhaps the best summary of how my feelings about irregularity have evolved comes from a meme that I saw on Tumblr many years ago. I have since seen quoted in screencap form all over the place, sometimes attributed to me, which I didn't come up with. We will link to the whole blog chain of the original source. Can you help me read this meme? Sure. When you don't know anything about linguistics. The plural of memorandum is memoranda. Why can't people get it right?

42:44When you know a little about linguistics. The plural of memorandum should just be memorandums because that's how people naturally say it. Memoranda is just prescriptivism. When you know a lot about linguistics. Oh my god. So certain English words were borrowed from Latin and Greek have competing plural forms with one form using the English plural S and the other using a borrowed Latin or Greek form. Do you realize how cool this is? A language borrowing inflectional morphology from another language. And here the two competing plural forms have become markers of education, expertise, social class. Isn't that incredible?

43:15When you have a degree in linguistics and DGAF anymore. Memorandables.

43:21Memorandables. For more Lingthusiasm and links to all the things mentioned in this episode, go to lingthusiasm.com. You can listen to us on all the podcast platforms or lingthusiasm.com. You can get transcripts of every episode at lingthusiasm.com slash transcripts. You can get linguistic zines at lingthusiasm.com slash zines. And you can follow at Lingthusiasm on all the social media sites. You can get merch with lots of linguistics designs on it, including our new and deeply cursed meat

43:55clarinet drawing, which also comes with its own mini zine at lingthusiasm.com slash merch. I can be found at gretchenmcculloch.com and I'm on social media as gretchenmcculloch.com on blue sky, at gretchen.mcculloch on Instagram. My blog is allthingslinguistic.com and my book about internet language is called Because Internet. My social media and blog is Superlinguo. Lingthusiasm is able to keep existing thanks to the support of our patrons. If you want to get an extra Lingthusiasm episode to listen every month, our entire archive of bonus episodes to listen to right now, or if you just want to

44:25help keep the show running ad-free, go to patreon.com slash lingthusiasm or follow the links from our website. Patrons can also get access to our Discord chat rooms to chat with other linguistics fans and be the first to find out about new merch and other announcements. Recent bonus topics include the linguistics of poetry, deleted scenes about Czech with Danny Bate and the Mysterious Voynich Manuscript with Claire Bowen and The Lingthusianist, a crossover chat with Helen Zaltzman of The Illusionist Podcast. Can't afford to pledge? That's okay too. We also really appreciate it if you can recommend

44:56Lingthusiasm to anyone in your life who's curious about language. Or leave us a nice review, like this one from Atcode, who said, learning and loving it. As someone that's always lay on the fringe of linguistics, using consciously my work but never quite understanding the building blocks of it all, this podcast has been invaluable. Great chemistry between the hosts sharing accessible and fun insights. I could not recommend it more strongly. Aww. Lingthusiasm is created and produced by Gretchen McCulloch and Lauren Gorn. Our senior producer is Claire Gorn. Our editorial producer is Sarah Doppiarella. Our production assistant is Martha Tsutsui-Billens. Our editorial assistant is John Crook. And our technical editor is Leah Vellman.

45:29Our music is Ancient City by The Triangles. Stay Lingthusiastic!