VoE - Björn Köhnlein === [00:00:00] Bjorn Kohnlein: One of the miracles of language is that we, as speakers of our languages, are somehow able to figure out what abstract categories someone else is saying, even though their speech is variable, gradient, right? Everyone has their own kind of tone, their own pitch height, their own pitch range, but we have some kind of mechanism that allows us to derive these abstract categories. Jen Farmer: From the heart of the Ohio State University on the Oval, this is Voices of Excellence from the College of Arts and Sciences, with your host David Staley. Voices focuses on the innovative work of Arts and Sciences faculty and staff. With departments as wide ranging as art, astronomy, chemistry and biochemistry, physics, emergent materials and mathematics, and languages, among many others, the college always has something exciting happening. Join us to find out what's new, now. David Staley: I am delighted to [00:01:00] welcome Björn Köhnlein into the Arts and Sciences Marketing and Communications Studio today. He is an Associate Professor of Linguistics at the Ohio State University College of the Arts and Sciences, where his core expertise is phonology and its interfaces with phonetics and morphology. Dr. Köhnlein, welcome to Voices. Bjorn Kohnlein: Thanks so much for having me, David, it's a, it's an honor to be here. David Staley: I'm very, very happy you're here. I'd like to start off with some definitions first before we dive into your work. So, I've introduced you as someone who's interested in phonology and its relationship to phonetics and morphology, and I thought maybe we'd start with some definitions: what is phonology, first of all? Bjorn Kohnlein: Yes, phonology, in the broadest sense, you can understand as the study of sound systems of languages, right? I mean, it's obvious that all languages have things like consonants and vowels, so all languages have that in common, you can also say all languages have things like words syllables, but there are of course also crucial differences between languages, right? So in Georgian, the claim is you can [00:02:00] start words with up to eight consonants, and you cannot do that in English, right? David Staley: I don't think I've ever heard that. Eight consonants? Bjorn Kohnlein: Yes. Yeah. I think it's eight. David Staley: Oh. Bjorn Kohnlein: Yes. But then the, the debates are, is that really one syllable or is one of these consonants sort of the core of one syllable and there is a split, so there are complications, but the point is you get sort of sequences of sounds in some languages that other languages do not allow, right? So, there are many languages that only allow sequences of like one consonant, one vowel, one consonant, one vowel, and English is sort of, let's say, in the middle of that kind of spectrum, right? In English you can get words like stream that start with s, t, and r, so you can have up to three consonants at the beginning of the words, but if you're interested in the system of sounds, you can already see one kind of idiosyncrasy there, which is that it has to be an s, if at the beginning of an English word you have three sound, three consonant sounds, it has to be an s. So, stream is an okay word, [00:03:00] but "sfream", the f and the s are not that different, but stream is completely out, right? So, that is one part of what phenologist are interested in, the sequences of sounds that are allowed or not allowed in the language, and the other part, and that is what I am mostly working on, are sort of all these things that seem to float above the individual sounds, that's something like syllables, right? If you have a word like wagon, that will be two syllables, right? Every young child can start clapping syllables very early. David Staley: Mm-hmm. Bjorn Kohnlein: So we do have intuitions for that. I mean, in English, there are complications when it comes to that, like for instance, people will debate whether fire is one or two syllables or not. Everything is always clear cut in that kind of realm, right, but we are interested in that, and then we have things like, like stress, like prominence or there, impact is a noun, impact is a verb. So, you can see that in English, this difference between stress on the first or stress on the second syllable can make a difference [00:04:00] between nouns and verbs. And then these groupings extend all the way up to things that we call phrases and intonational markings, right? So when I give you a sentence, " this is a house", you can hear, okay, i'm making a statement about some object in the world, in this case, a house. When I say "this is a house?", it has the exact same kind of word order, it has the same words, but everyone can hear, okay, this is not someone making a statement that is someone maybe asking a clarification or a surprise question, right? David Staley: Mm-hmm. Bjorn Kohnlein: So those are some kinds of speech melodies that we have that we call intonation that we put on top of the utterances as we speak, and those are things that we generate, necessarily, on the fly because we say many sentences that we have never said before and that maybe no one has ever said before, and they always have those melodies. David Staley: So, phonetics and morphology. Bjorn Kohnlein: Mm-hmm. David Staley: So phonetics, I know from like what little kids learn, but that's... Bjorn Kohnlein: Yes. David Staley: That's not the same thing that you study, or...? Bjorn Kohnlein: Well, [00:05:00] phonetics and phonology are intimately related in the sense that , phonology, we think of as sort of these abstract categories, you know? There is a t and there is a d. However, like the difference between t and d is one of what we call voicing, that sort of vibrations of the, of the glottis. Some other aspects play a role as well, but the point is those things are gradient. That's what phonetics is, right? So, when you say a million ts or a million ds, you will actually never produce the exact same sound twice, and yet we are able to categorize them, right? You can do that with words. If you say house,house,house,house, and you would really measure that, you could see now I said four times house, but all of those are different. David Staley: Hmm. Bjorn Kohnlein: So in that sense, phonetics is gradient, but it is intimately linked to phonological categories, and one of the miracles of language is that we, as speakers of our languages, are somehow able to, [00:06:00] figure out what abstract categories someone else is saying, even though their speech is variable, gradient, right? Everyone has their own kind of tone, their own pitch height, their own pitch range, but we have some kind of mechanism that allows us to derive these abstract categories. David Staley: Hmm. You just used the term speech melodies. Bjorn Kohnlein: Mm-hmm. David Staley: And I'm very interested to know what that means, because when I hear the term melody, I think music. Bjorn Kohnlein: Mm-hmm. David Staley: And I don't know if I've ever heard a linguist use this term. What do you mean by speech melodies? Bjorn Kohnlein: Yeah. It is like, I'm coming back to that example that I, I used before, right, "is this a house?" versus "this is a house", right? In, " is this a house?", you sort of have some kind of, what we call rising tone at the end of the sentence on house, so your voice starts at a low level and you go up, right, versus in "this is a house", you have the same, just it's the opposite melody, you go from a high pitch to a low pitch, and many linguists assume that these speech melodies, which we use to convey linguistic [00:07:00] categories like, a question or yes, no question versus a statement. There's also a difference between if I say, I'm gonna tell you a story, and now I stop and it's weird, because I said I'm gonna tell you a story and you expected a continuation. David Staley: Yeah. Bjorn Kohnlein: Based on me staying at a high, sort of high level pitch at the end versus if I'm gonna say, I'm gonna tell you a story, then you're sort of, more happy with me taking a break, you think, okay, probably they're just recollecting themselves and the story will start, but if I end on that high thing, there is sort of is... I am holding the turn, I'm continuing to speak and we all have some kind of categorical knowledge of that. Of course, we can all have misunderstandings because of intonation, I think everyone who's ever had a relationship knows these kinds of, you sound annoyed, no, I don't, why, you sound annoyed. So that, that just shows that it's a complex phenomenon because emotions can come into play, you know, some people have a high pitch range and sound very excited, other [00:08:00] people speak more mellow, and then the pitch ranges are not that wide. But one crucial aspect is that we need to be able to figure out, okay, this is a, say, yes, no question versus a statement, right? And so, speech melodies in that sense, many linguists use actually tones as sort of anchor points to define where those kinds of rises and falls start. Typically, that's unsurprising, that is the most important element of a sentence, and that's usually towards the end, right? So in, "this is a house?", house will be the most important element, and then you link, you sort of anchor the relevant tones there, say if it's a low high, it will give you a question, if it's a high low, then it will give you a statement, right? There are some more complications there, but that sort of is the baseline ,and what exactly the nature of these tones is and how many we need to distinguish, there is variation and disagreement because it's just a hard field to [00:09:00] study, you know, because in something, if you wanna study the difference between house and mouse, you have like a minimal pair, one starts with an h, one starts with an m; but in these intonational categories where so many factors come in together, linguistic factors, emotional factors, you don't get as easily things as minimal pairs and you get a lot of variation, but sort of, that's the baseline. People use these international tones as anchor points, and you put that on top of your sentences and then sort of, that's what the phonology still does, and then the phonetics gets some kind of manual and reads off of that. David Staley: That rising intonation. Bjorn Kohnlein: Mm-hmm. David Staley: That, that signals question. Is that true across languages? That's true in English, is that true in, I'll say German? Is it true in Mandarin? Bjorn Kohnlein: Yeah. That's, that's a very interesting question. People have thought for a long time that that is the case, at least it was sort of in the early years of intonation research, it was around as an idea that rising intonation is tied to questions and falling into nation is tied [00:10:00] to statements. However, we know by now that this is not, it's not true. Like, for instance, I taught at a graduate school in Ghana, like around 15 years ago, and one of the many languages that people speak in Ghana is Akan, it's a tone language, and I remember that I had read that what they do in Akan is marking questions with a low tone at the end, right? So, I recorded some of the people there because it was curious and it was, I think, the sentence was sort of Kofi, which was a name, Kofi is coming or going somewhere, just that short sentence, right? So when you say that as a statement, and I, okay, my Akan is terrible, I'll still say it the way I remember it, it was something like "Kofi kɔ". That was a statement, and that just has the so-called lexical tones, right? So there are some languages that use tone lexically to distinguish word meanings, and Akan is one of them, and you add that question tone at the end if you ask a question and then it becomes "Kofi kɔ?". So, [00:11:00] you go from high to low, but "Kofi kɔ" in Akan is a statement, "Kofi kɔ?" is a a question, hence it is not a universal, but it is a tendency, right? David Staley: Hmm. Bjorn Kohnlein: But you can even see, if you just think about West Germanic languages, right? So, those are, English is a West Germanic language, and it's, closely related to Dutch, which is closely related to Afrikaans and German, there's also Frisian. In various varieties across that area, what you do find is actually you do have dialects that use rising intonation also to mark statements, right? Which, then, for speakers of varieties of West Germanic that are not used to that, it can sound like those people are asking questions all the time. Maybe some people who may have seen _Norsemen,_ it's a Netflix comedy show. It's a Norwegian show, but they filmed it in Norwegian and they've also filmed it in English, but some of the actors, they have a, sort of urban East [00:12:00] Norwegian intonation, which sort of rises at the end of phrases, you know, and so you can actually hear there, there is something very unique about it, and people in Switzerland, and there are so, so-called Alemannic dialects, there are also many dialects where the most prominent syllables, which usually have a high tone and then fall in statements, have a low tone and then rise. So, it's not a universal, even in the languages that are very close to English. David Staley: When we say melodies or when linguists say speech melodies, is that simply an analogy or is there any sort of crosswork with music or music theory? Bjorn Kohnlein: There are definitely many similarities, right? I would say that the use of pitch in singing and speaking is, I mean, it's obviously not the same, right, because you can hear when someone sings, but when they speak it's different. So in singing, I mean, I'm not like an expert on those direct comparisons in [00:13:00] melody, but you find, for instance, like notes that are much longer and stay level; you do not find that that often in speech, so there's more up and down movement maybe in kind of shorter terms, and melodies in singing are much more variable than melodies in speech. You know, in speech, what you really typically find is at least in so-called intonational languages like English, Dutch, Afrikaans, German, you go up and you go down throughout the sentence, and there are certain anchor points where that changes, right? But there definitely are similarities there. There are also obvious similarities in the, kind of, what we call like metrical organization of languages like, you know, prominence and rhythm. Even linguistics uses, so in poetry, for instance, right, and also in in music and lyrics, you have like this notion of like a trochaic and iambic feet, when you have something like ba ba, ba ba, ba, ba, right? So, initial prominence and then prominent, non prominent, prominent, non [00:14:00] prominent or strong, weak, that's a trochaic pattern, but when you have da, da, da, da da da da, that's sort of an, a kind of iambic pattern, right? In poetry, there are many more verse meters than that, but languages seem to use troches and iams, right? There are some languages that have like ternary stress patterns where you get something like ba bup bup, ba bup bup, but those are usually more complex, there are like certain kinds of possible interruptions, but the point is you can divide the set of languages that have some kind of word level prominence pretty well into trochaic and iambic languages, which in and of itself, I think, is a fascinating property of language, but which also shows sort of the similarities between these kinds of different domains it's as if in poetry we've taken something that, who knows, maybe was there in language already and then we've, we've expanded on it, you know? But you can also see it in the terminology, like [00:15:00] people in linguistics use feet for that which is of course also what you use in poetry. David Staley: In poetry, yeah. Bjorn Kohnlein: Right. So, English is typically assumed to have trochaic foot structure, which means that, you know, like most English native words, I mean, they're so short that there is not much rhythm there, right, because if you think about your body parts, like eye, nose, mouth, chin, ear, head, hair, well, liver has actually the two syllables, but what you typically have in the most basic vocabulary are one syllable or two syllables, and then a second syllable has like this reduced vowel, like a schwa. So, there's this kind of preferred template, which is like, either it's a monosyllabic foot, then of course you could call it anything, right? If it's a disyllabic foot, it's a disyllabic troche. David Staley: Let's, uh, talk about some specific work you've done recently and you've been studying intonation patterns of early Germanic languages. Tell us about this research. Bjorn Kohnlein: Yes. I will say that, I mean, there are all these quotes around that basically say that it's, almost [00:16:00] hopeless to study the intonation of early languages, and I actually do agree with those kinds of people because, I mean, early Germanic is its own kind of thing because, you know, like the Romans and the Greek, they all had all their philosophies stuff and histories, you know that, right, a lot of it written up; the Germanic people not really, you know. There's like very, very little evidence. There's like a gothic Bible, fourth century, I believe, common era, and that is basically what, that's the core of our understanding of early Germanic. I remember a few years ago they found a comb somewhere in Germany from, I think also like the third, fourth century common era, and it was a spectacular find because it had the runes for the word comb on it. And that was like really good evidence for the use of runes at the time, but the point is, it's one word on a comb, and it's a spectacular find for like [00:17:00] hundreds of years into the common era, right, where we had all the Greek and Roman philosophers. So, that means that early Germanic is hard anyway, and another problem is that when you get things like that, like modern orthography doesn't give you any kind of indication of, well, intonation sentence melodies, you know? You get an exclamation point, you get a full stop, you get a question mark, you get a comma; how those are pronounced, I mean, we do have an idea, but if there were no oral records of our speech and then, you know, in 2000 years people would find all kinds of internet things, if they wouldn't include commentary on that, they would have no idea how to pronounce it. Even with commentary, it can be hard, right? So, that is one thing, we just do not have any direct evidence whatsoever. There are some like older texts that have markings. For instance, there were like two authors in old High German, Notker and Otfrid, who did put some kind of marks on that, and it seems pretty clear that that was supposed to mean some kind of intended [00:18:00] pronunciation, but it's completely unclear, ultimately, what they mean and whether they even relate to anything specifically melodic or whether they just mark, Hey, this is a prominent word, this is not as prominent words. So that's the challenge, it starts with even trying to do something like that. David Staley: Before we started recording, you'd said that this summer you're going to be going to Germany and doing field work. Bjorn Kohnlein: Yes. David Staley: I'm interested to know what field work means for a linguist. Bjorn Kohnlein: Mm-hmm. David Staley: What does that mean, and I guess what sort of methods do you use to study language? Bjorn Kohnlein: Mm-hmm. Language can be, I will say this with regard to like phonology, right, and there are many sources of data, and I don't think it would be fair to say that there is like one source of data that, out of principal, should be preferred over another, right? So, the traditional approach has been to rely on your own intuitions about your language and then to write something about that up. That, by now, is often discarded [00:19:00] for maybe good reasons, because people, they can make themselves believe all kinds of things, right? I notice that myself if I, if I wonder about something in German, if I wonder about it, like 10 times, 15 times, at some point, I have no idea anymore, right? There are challenges to introspection, but still it has yielded very interesting and fruitful results, right? Other ways could be just, you know, informal conversations with native speakers, asking them about their intuitions, which sort of already detaches you as the researcher from the object of study a little more, but, you know, people don't necessarily give you an answer as to what the grammar of a language is, right? I mean, for instance, there's a reason we go to the doctor when it hurts somewhere. So, we know that it hurts, but we don't necessarily know why, right? So in that sense, uh, when you ask someone, okay, can you say this? No, I can't, they, maybe they, if you push them, it can say, well, maybe it's because of that. They might be right, they might be wrong. I've talked to some of my informants in the past, so what do you think this is, and they said, oh, it's [00:20:00] that and that, and then experiments turned out to show that actually, no, it's not that, you know, so that's, when asking people, people directly, that's a very important source to understand the language, but it comes with downsides. Then, you can use sort of, what I often do is use more targeted kind of sentences that aim at eliciting, in my case, certain words and certain positions in the sentence, because I wanna find out how long are the vowels and what are indeed the melodies those words are produced with. That is nice because it's controlled, right, and it can give you things like minimal pairs, where you see one word has that property and the other word is only different in that other property, you know, so, and then you can sort of measure that, you can see there's a significant difference. You can play those things back to people, you can see whether they can get the checkbox, yes, you know, this is a basket, yes, this word means man, or something like that. You can manipulate those words to figure out, when does the perception switch and what other things that aid this [00:21:00] perception? And then, another approach is to use corpus data, that's becoming increasingly popular, right, which is fine, you know, I mean, the downside of course, of controlled speech, you can say that's not how we speak in everyday life, yeah, that is true. But I mean, one, you can, you can wonder, like, why would people have like some kind of studio speech where they produce things that otherwise they would not produce? David Staley: Like what we're doing right now? Bjorn Kohnlein: Yeah. For instance, right? So I don't imagine that a baby in a German village that was born in 1935 acquires some kind of language that it will use if it ever is in a formal setting, where it is experimentally investigated, but corpora are very useful because we, I mean, one thing that we can see is that there is lots of variation, right. David Staley: Maybe tell us what corpora or corpus refers to? Bjorn Kohnlein: Yes. Okay. Corpus refers to, like, for instance, here we have the Buckeye Corpus of Spoken American English, and that is basically, you know, you take language data that have been recorded in various kinds of contexts, that can be free speech [00:22:00] that was recorded somewhere, it can be radio, it can be lots of different things, right, and, and there can be many sources for those kinds of things that can be recorded, that can be created in a targeted way, or maybe it's things that are lying around somewhere, you know, that have been recorded in present times or a long time ago for completely different purposes. So, you can use those, to, let's say, annotate them. You know, you can annotate what words are in there, you can annotate what consonants are in there, what vowels are in there. You can do a lot, right? It's a lot of work, which can, in modern times, some of that can be automated in very useful ways, but still it has to be checked, right, because there can always be mistakes. But that sort of gives you a window into speech where, ideally, there's like no one telling someone, Hey, can you read that sentence, you know, so you can say that's the most natural that you can get, and maybe, unsurprisingly, that's also where you find the most variation, right, which is one of these areas [00:23:00] where, sort of the more formal approaches to linguistics or to phonology have come under attack because all of a sudden, you know, like 20, 25 years ago we could see, oh, there's so much variation there, and maybe all those categories aren't really there in the way that people assume they are. David Staley: Hmm. I'm curious to know, did you always have an interest in languages? Maybe when you were a child, you didn't think, well, I'm gonna be a linguist, but did you always have an interest in language, or is that something that developed later as you got to university or graduate school or...? Bjorn Kohnlein: I did not intrinsically get to the point where I realized how fascinating it would be to study language. It was a complete coincidence almost, right? David Staley: Mm-hmm. Bjorn Kohnlein: Because at first, I mean, my sister had studied medicine, so I thought, okay, I wanna do something else 'cause I don't want to do the same, and then, you know, like late high school, I was like, eh, okay, let's try psychology. So, I tried psychology and I mean, it was great, but the part I liked most [00:24:00] about it was actually sort of everything that had to do with language, so that was sort of the first step. Then, I moved towards like the more sort of language specific domains, but I also wanted to pursue this idea of media studies, you know, because around like the year 2000 ,TV and radio and newspapers that all were like still big things and it was really cool and everyone wanted to do that, so I was like, ah, why not, I'll do that and I'll do some like, sort of German language and literature on the side. And then I went into my introduction to like linguistics class, and that was the first real introduction, I mean, psychology, I heard a little bit about psychological experiments on language, but in that class I learned for the first time about abstract patterns in language, and that is sort of what sparked my interest. I also did well in that class, and then they were asking me whether I would want to become a undergraduate research assistant there, so this is how I got to that, right? And I liked it, it was great, but I also did sports journalism on the [00:25:00] side, and then, came sort of the moment where I had to decide, do I want to do sports journalism or do I want to do a PhD in linguistics? And I had this opportunity to go to the Netherlands, to pursue a PhD, because I, they had a project I had applied and they had offered me the job. So, after some back and forth, I decided, yeah, I'll do the PhD, and that's when I learned about, like, what people call generative grammar, and that changed everything. David Staley: You should probably define for us what generative grammar is. Bjorn Kohnlein: Yes. Generative grammar is, i'll try to be very informal here. Right, it is sort of the idea that we have some kind of innate competence to learn language, right? So, that means, as babies, we are born with some kind of Lego set of possible patterns in language, and what the child is doing, it sort of puts them together to create the kind of system, which is the language that it grows up with, right? [00:26:00] Because, I mean, what we know is that languages, themselves, are not genetic. If a child is born here and in its first few months, the family goes to live in China, only speak Mandarin, let's say it's American parents, you know, and, but they all only speak Mandarin there hypothetically. The point is the child will learn Mandarin, will not learn English. It will not have a more difficult time learning Mandarin than a child who has Chinese parents and was born in Beijing, right? So the point is, languages, like English and German, they do not have any genetic basis, but the assumption in generative grammar is that at least parts of how we learn language are informed by these kinds of building blocks that we have underlyingly, and those are two parts, those are, let's say like categories, representations for phonology, that would be indeed, that there are consonants, that there are vowels, that that can be tones, that there can be stress so, all of those kinds of things the child knows and then puzzles together what its specific language is. And on the other hand, there's a thing such as [00:27:00] grammar, you know, so if you think of the difference between to write, I'm overdoing it here, a letter or to ride a bike, right? Those are different; one ends in a t, the other one in a d. But when you say, he's a writer, it could actually mean both; it could be someone who writes books, it could be someone ,you know, who is out on their Harley all the time. And that is because the t and the d in this specific position in the middle of a word, tends to become this, what we call the flap, this duh, writer, you know, and then contrast can go away based on some kind of language rules that children learn without being taught, that most adults are never aware of. David Staley: Hmm. And we know that generative grammar has been criticized. Bjorn Kohnlein: Yes. David Staley: But you made an interesting statement about it. Well, what's your attitude, I guess, toward generative grammar? Bjorn Kohnlein: I mean, first and foremost, I think in the field of linguistics, like many of these debates over the past few decades have become a little too religious, you know. David Staley: Because this has been an important [00:28:00] part of linguistics since the fifties, right? Generative grammar, yeah. Bjorn Kohnlein: Yes, it is. It is, right. But, I mean, it's important that we have some kind of theory that allows us to make statements about the object, right? Whether generative grammar is the right one or the wrong one, or whether those are even the right terms to talk about things like that, is a different question, but what it does is that it gives you a set of axioms, right, that you use to look at empirical data from a variety of languages, and because generative grammar is thought of as something universal, right, it's a universal capacity that we all have; that, by necessity, would imply that all languages are typologically related, they are a set where you can compare individual members to each other, and where that comparison will yield insights not only about what we find across languages, but ideally also some kind of insights about what we do [00:29:00] not find, right? If you have generative grammar, you make implicit, explicit claims about what are possible languages and what should be impossible languages. But that is an empirical inquiry, you know? So, you can have a great theory, but then there is like one language that does it differently, and it doesn't matter whether it's one or whether it's seven hundred, right? The fact is it is a language that people can learn, hence, the theory has to be adapted. It either has to be discarded or it has to be modified, and I've found for myself, I found this a very satisfying approach to try and make sense of patterns that I see. Sometimes when you're lucky, you even see patterns in a language that you've never thought before, and you sort of see, oh wait, this looks like that, and that, you know, like sometimes because of abstract kind of generalizations that you have made before, and I find that aspect very satisfying. David Staley: Hmm. Tell us what's next for your research. Bjorn Kohnlein: Yeah. I mean, I always have multiple projects [00:30:00] at the same time. One is this kind of reconstruction of a Germanic intonation that I touched on before, right, which you can do via comparing languages, I mentioned the Norwegian English and the Alemannic, and of course Norwegian varieties. Another thing, like maybe a bit more of a quirky thing, which I do think can inform our understanding of sound systems and of stress and things like that, is this thing that's called explative infixation in English. So, it is where you like, insert a swear word into another word. So, I'll just... David Staley: Do you wanna try an example? Bjorn Kohnlein: Yeah, sure. I'll, I'll just go for it, you know? So, a classic is abso-f***ing-lutely Bjorn Kohnlein: you know? So, the point is, everyone, I guess most people know that, some people like it, some people don't. But if we abstract away from that, the idea is that it is usually said, and almost you could say, has to be said abso-f***ing-lutely Bjorn Kohnlein: you know, so it goes exactly there, and one of the questions, one might ask is, why actually, why does it go there? You know, if you say [00:31:00] absolute-f***ing-ly, Bjorn Kohnlein: it sounds very weird, right? David Staley: Yes. Why does it sound weird? Bjorn Kohnlein: Yes, exactly. One of the things that we know about it is that it comes preferably before a stressed syllable. So, if you say absolutely, you know, then "lute" is the main stress syllable in absolutely, so that is sort of your target for this kind of in fixation. And if you say absolute-f***ing-ly, Bjorn Kohnlein: it actually comes after the stress syllable before an unstressed syllable, and that's really weird. So, I wanna do an experiment and I'm working with three undergraduate students from our linguistics department on it, where we sort of also want to test like how bad certain positions can be, so it's, for instance, absolute-f***ing-ly Bjorn Kohnlein: I think we can agree is bad, right, Bjorn Kohnlein: ab-f***ing-solutely is also what no one would say, but the question is, is it equally bad as absolute-f***ing-ly? Bjorn Kohnlein: And, at least we do think maybe it is not, because [00:32:00] at least the stress syllable is still coming, sort of, you know, Bjorn Kohnlein: ab-f***ing-solutely flows better than absolute-f***ing-ly. Bjorn Kohnlein: But, interestingly then there are also words, like in a word like Google, you just cannot put it; Bjorn Kohnlein: Goo-f***ing-gle sounds Bjorn Kohnlein: terri-f***ing-ble, you know? So, no, that's, so there are like different types of words, different types of constructions where it fits in better than in others, and there is some kind of underlying mechanism, and I, myself, I am interested in those kinds of cases that are bad, but maybe there are different degrees of bad, you know, and then you can do that with, you can do it with place names, you can do it with nouns, you know, Bjorn Kohnlein: o-f***ing-hio is pretty okay, right, Bjorn Kohnlein: Ala-f***ing-bama, but, Alaba-f***ing-ma Bjorn Kohnlein: is really weird, right, Bjorn Kohnlein: A-f***ing-labama. So you can, you can manipulate all those kinds of things in a way, and then what you can do, for instance, is the most straightforward way is just ask [00:33:00] people for their judgments, how good does this sound? And I do believe, saying this at a very high level, that interesting insights about how English words are structured can emerge even from variation within those positions where it is bad. What is the worst, what is less worse, right? So one example, a last example that I want to give to sort of indicate that the power that the stress has to shift, right? So, you have words like alligator, where the main prominence is on alligator, you know, but if you say alli-f***ing-gator Bjorn Kohnlein: it has to go to gator. You know, Bjorn Kohnlein: alli-f***ing-gator, you cannot say alli-f***ing-gator Bjorn Kohnlein: I mean, you can say it, but it sounds weird. You say alligator, but Bjorn Kohnlein: alli-fucking-gator. So, there you see the power of that infix, that it can even, under certain circumstances, manipulate the position of stress within a word. David Staley: Hmm. Björn Köhnlein, thank you. Bjorn Kohnlein: Thank you so much for having me. Jen Farmer: Voices of Excellence is produced and recorded at the Ohio State University [00:34:00] College of Arts and Sciences Marketing and Communications Studio. More information about the podcast and our guests can be found at go.osu.edu/voices. Voices of Excellence is produced by Doug Dangler. I'm Jen Farmer.