What's in a word? An introduction to morphology

[Music] Hello and welcome to episode three of the linguistics 8 key podcasts. In episode two I gave you an introduction to phonology or sound systems and I introduced the concept of the phoneme which is the smallest meaningful unit of spoken language. Today we're going to look at morphology which is the study of how those units can be put together to make out words. First of all we need to think about what a word actually is and here there are just as many questions as answers I'm afraid. The colonstictionary defines a word as a single unit of language that can be represented in writing or speech. The merion web sedictionary calls it a sound or combination of sounds that has a meaning and a spoken or written and according to the Oxford linguistics dictionary it's the smallest of the units that can make up a sentence. In addition the American linguist Leonard Bloomfield defined a word as a minimal free form. The difficulty with Bloomfield's definition is that it could be interpreted as meaning that individual words can occur in isolation but mostly they can't. For example we wouldn't usually utter words like 'went' or 'at' or 'of' on their own. Also compound words such as clickbait or sleepwalk are considered to be words but they aren't minimal free forms as they're made up of two words joined together. One word isn't necessarily equivalent to one unit of meaning and this is where cross-linguistic differences come into play. For example 'master' or the legsyles of Vopin is one word in Norwegian but four words in English namely 'weapons of master's traction'. In a glutinating language is like 'now I've spoken in Mexico'. Several grammatical elements can be joined together to make one word rather than a whole sentence such as in the word 'nemitspanos' meaning 'I will pass you' which is certainly as a way of greeting someone when you don't have time to talk to them. In languages with grammatical case a word can occur in many different forms according to its function in the sentence. The Slovenian word 'dan' for example meaning 'day' can have 16 different forms. So should we consider them to be 16 different words or should we just consider it as one lemma which is the term we give to a dictionary form or base form with 16 variants? We've also got the problem of homophones namely elements which look like distinct words in writing but sound the same. Does it make sense to class them as distinct words if we can't always tell them apart? How many different words for example do you think there are in this French sentence? Le ver, ver, ver, ver, le ver, on ver? This is about the green worm going towards the glass made of fur incidentally. Well it sounds in the French sentence as if it consists of five different words but actually there are eight and then we've got the matter of sign language. The definitions we've considered so far have mostly referred to sounds and speech but what about languages which don't use sounds and speech? Can they also be considered to have words as meaningful units? Another question that often arises when talking about words is how many a speaker actually knows. The number of words a person knows depends really on our definition of a word which as we've established is not very satisfactory and on whether a foreign borrowing can be considered as one of our words or whether different parts of a word such as 'ran' and 'run' counters two words or one. It also depends on a person's degree of language input and their age. Research suggests that a 20-year-old American might know 42,000 lemurs but it could vary from 27,000 to 52,000 depending on the person. And many of the words that the person knows may just be passive knowledge in that they might be words that the person understands but wouldn't really use themselves in a conversation. People who are learning foreign languages often wonder how many words they need to know and again it depends entirely on the subject matter and the purpose for which you're learning the language. Do you want to be able to read foreign novels? Do you want to be able to watch a TV program? Or do you want to be able to communicate on a basic level when you go abroad? Exam boards in Britain at least don't publish specific word counts that students have to learn. In the common European framework of reference for languages, the C2 level, which is the level at which a person can understand virtually everything and can communicate on fairly complex subjects, a vocabulary of 10,000 active words to 20,000 passive words is assumed. People often wonder as well which language has the most envious words when it depends as vocabulary is infinite and there's no real way of counting exactly how many words a language has. It's often said that there are certain languages which have numerous words for snow or sand or some other element which is very prevalent in the environment or culture where that language is spoken. This is called hypersenonomy and it is true that Somali, for example, has a large number of words for camel with very varying degrees of usage, including Arran, a camel that has been weaned or Abir, a female camel that hasn't given birth. The semi-language spoken in northern Finland, Sweden and Norway, has about a thousand words relating to reindeer, such as Leami, which is a short, fat female deer, or Snorri, a deer with short, branched antlers. In urban British English, there is a large range of sign words for guns and knives, including Berner, Skeng, Spinner, Shank or Ramzi. As is the case with sign words, these are constantly changing with new ones being introduced and others losing popularity. Normally a word has to be documented in a dictionary for it to actually count. It's a bit like world sports records where the record holders are those who have been observed running or jumping under controlled conditions, while there may well be other completely unknown individuals in far-flung corners of the world who could easily outdo them. There are a number of processes at work in the creation of neologisms or new words, and I'll go into more detail about those in the episodes on semantics, contact phenomena and sociolinguistics. For now, we're going to think about morphology and how words are put together using methods like compounding acronyms, blending and affixation. So just as a phoneme is a minimum unit of meaningful sound, a morpheme is a minimum unit of grammar, and there are various different types of morpheme. We've got three morphemes which can exist by themselves, such as raccoon or hungry or is. They can't be broken down into separate components. They're free sending elements on their own, and they don't necessarily need anything else to be added to them to be able to appear in a sentence. The large raccoon is hungry. It's a sentence made up entirely of free morphemes, for example. We can compare these with bound morphemes, which can't appear on their own, but only as part of another word. Elements like "un" or "ree" or "lie" as in "unkind" or "revisit" or "quickly". Interestingly, a bound morpheme, which has started to appear on its own, is "ish" as in "greenish" or "toolish". For example, some of my tasks did the interview go okay, and you might reply "hmm, "ish". We've also got inflectional morphemes. These are bound morphemes which add new meanings to an existing word, to give a different version of that word, without changing its function in the sentence, such as the plural "s" in "shop" "shops", and the past tense suffix "ed" as in "walk" "walked". Essentially, inflectional morphemes tend to be grammatical. They give us information about an existing word. Is it a past or present verb tense, for example? Is it a plural or singular noun? Is it a feminine or a masculine noun in languages that make that distinction? Derivational morphemes, on the other hand, abound morphemes which, when added to a word, can change its meaning or its grammatical category. For example, adding "ness" to "happy", an adjective creates "happiness" a noun, or adding "un-to-kind", creates "unkind", changing the meaning from "positive" to "negative". You may remember in episode one, we talked about "alaphones", which are the contextual variants of a phoneme. We've also got "alomorphs", which are variants of a morphem, and they give the same information, but they do so in a slightly different way, according to the context in which we find them. A good example is the plural "s" in English, which can be a "z" sound like in beds, and "iz" sound like in houses, or a "s" sound like in ships. We might say that this "alaphones" phonologically conditioned, because its pronunciation depends on the adjacent phonemes. When it's after a voice sound like "d", it's also voiced, which gives "z" as in beds. And when it comes after an unvoiced sound like "pur" as in "shop", it's an unvoiced "z". An "alomorph" can also be lexically conditioned. An example of a lexically conditioned "alomorph" is what we usually refer to as an irregular plural, for example, in English, such as "children" or "feet". They function as "plurals" in the same way as the "s" in ships or beds, but they don't follow any specific rule, and they have to be learnt separately. We can say that the word "feet" in fact represents two more themes, "foot" plus a plural. Let's look at the sum of the ways in which we can create new words from existing more themes. Affixation is where we add a prefix or a suffix to a word, which may change the word category. It might change it from a noun to an adjective, for example, or it might remain in the same category. Common prefixes include "anti", "re", "in", "pre", "un", "x", "mega", "uber", and they're not all native to English, so we don't just borrow a whole word sometimes, to talk about new concepts, but we borrow bits of words that we can then attach to our own words. And that's something I think is very interesting about English in particular. Some suffixes that we use in English include "ti-o-n" as a nation, or "hood" or "ism" or "t" or "ness" or "dom". "Dom" is a very old suffix dating back to old English. It's generally not very productive, but we do find it in words like "officialdom" and "singledom". In English too, we get some interesting mixtures of romance and Germanic elements in words such as "ungraciousness", where we have a Germanic prefix, a romance word in the middle, and then a Germanic suffix, or adulthood, a mixture of a romance word and a Germanic suffix, and then dwarfism, which is the other way around, a Germanic word, with a suffix that was originally Greek, and a apprenticeship, which is romance and Germanic. Some affixes are much more productive than art, such as "er" "er", to denote a person who does something by changing a verb into a noun, such as "bake", "baker", "right", "writer", which has been extended to "youtube", "youtuber", which are both nouns. Others like "en", "en", are not productive, such as in "titan", "hasen", but they are probably more productive in the past, given that they have related words such as "tight" and "hase". Interestingly, the grammatically correct way to say that you are removing someone from your list of contacts on social media is to de-friend them, as de-relates to the idea of removal, although the more frequently occurring "unprefixes" preferred among speakers as an "unfriend", which actually was the Oxford Dictionary's word of the year in 2009. Some languages take affixation to a bit of an extreme, in the case of a duplication, which is a feature of languages such as "Indonesian" and "Hawaiian", and happens when you repeat part of a word or a whole word to change the meaning slightly. For example, to form a plural in "Indonesian", you repeat the whole word, so "orang" meaning "person" becomes "orang" or "ang". Or you can use a duplication to change the word class, such as by making the adjective "diam" meaning "silent" into an adverb "silently" by changing it to "diam" "diam". The search on English has shown that when an adult reduplicates when talking to a baby, it helps the baby remember the word more easily, such as in "bye-bye", "too-too" and "dada". Compounding involves forming a word from two or more units, which are words in their own right, and this is a very productive process in Germanic languages, such as English, Dutch and German, giving us new words like "randsomware", "crowdfunding" and "deepfig". German and Dutch, in particular, are famous for their very long words. In principle, an infinite number of words can be joined together to make one compound. Once a Dutch TV channel invited viewers to send in the longest Dutch words that they could think of, and the winner sent in a word 60 letters long, namely "kinder kan aval subtocht for beriding's ver examin' komitilin" meaning children's carnival parade preparation work committee members, and I'm a superlgeist for my Dutch pronunciation. Personally, I find acronyms particularly interesting. Acronyms are words formed from the initial letters of two or more words. For example, "aewol", "absent without leave", "nato", "northatlantic treaty organization", or "fomo", "fear of missing out". I like acronyms because very often they become lexicalized or regarded as words in their own right, and people very often forget what the letters originally stood for, and often have no idea that they're dealing with an acronym. We hear a lot about care packages in our troubled times, which in fact started out in the 1940s as food parcels sent from the United States to Europe by the Cooperative for American Remittances to Europe, which has subsequently renamed the Cooperative for Assistance and Relief Everywhere, which of course spells care. Acronyms sometimes become productive, and they can undergo inflection and derivation and generate different word types, so they're not acronyms anymore, but free-sounding words. Quite a lot of acronyms in English go under the radar in this way, and that's a pun, by the way, because radar is an acronym itself, sounding for radio detection and ranging. For example, MC, Master of Ceremonies, has become a verb, as in, he earns extra money, I'm seeing in clubs. Loll, laugh out loud, can be pluralized, as in she did it for the lolls, and one of my favorites is the Tom A. Swift Electric Rifle, which is a taser, and this is now also spawned verb as into taser somebody. Acronyms can result in redundancy, however, when we find utterances like, please RSVP by Monday. Please is already encoded, rip on this simply, so we don't really need to say please again. And you do also hear people sometimes referring to pin numbers, and of course a pin is a personal identification number, so we don't really need to say the word number twice. Now, I believe at the end of the last episode, I promised you designer dogs today, and this is where we come to blending. Blending is when we take the first part of one word, and the last part of another word, and we join them together. And we've got boring things like smog made of smoke and fog, but personally, I prefer to regard blending as maths with dogs, which is much more fun. Recent years, I've seen a proliferation in hybrid dogs, and their names are almost always morphological blends. For example, springer spaniel plus coca spaniel equals sprocker. I wonder if the dog lovers among you can work out what these dogs are. We've got the chug, the whoodle, the puddle, the shorky, the pomski, and the kavashon. One which goes against the grain, rather, is the cockapoo, namely the coca spaniel and the poodle, which technically I think should be cockadoodle, but that's clearly a different beast all together. Clipping or truncation is one of the easiest ways of creating a new word, and it's where we remove part of the word, usually the end. So we get, for example, gymnasium, which becomes gym and advertisement, which becomes advertile ad. We might sometimes remove the beginning, such as telephone giving us phone, but this is less common. In the word fridge, we've removed both the beginning and the end of the word refrigerator. Clipping is different from abbreviation, in that an abbreviation tends to be found in writing, such as the use of the letters d e p t to sound for department or a v e for avenue, whereas truncation or clipping relates to the formation of a freestanding word, which can be spoken. And interestingly, just like with acronyms, words resulting from clipping can be productive. They can generate new word types. For example, from bicycle, we get bike and then biker or biking. We can make compounds, for example, infographic, webcam and chatbot. And then we've got gastropub, which is a clipped form of pub from public house together with gastro, a learned prefix originally from Greek, which is mostly used in a medical context. Incidentally, gastropub is something that the French are not particularly keen on, given that gastro is their term for a stomach upset, yikes. Anyway, finally, we come to back formation, which is affixation in reverse. Namely, you remove an affix, which is part of an older word, and you create a new word. Most latent rules creating a verb from a noun, although very often, we tend to assume that the verb came first. For example, edit came into English as a back formation of editor, and we also got televised in the existing word television. This process could often result in redundant words when a noun and verb pair already exists, such as obligate from obligation, when we already had a blight, or predate, which arose in the 20th century from predator, and now exists alongside the word prey, which came into English in Norman times. So in this episode, we've seen how words themselves are built, but what happens when we start swinging them together? Why does word order matter? And how do different languages play by different rules? And what's the real story behind headlines like schoolboy helps dogabyte victim, and border force can't stop smuggling? That's the power of syntax, and we'll be diving into it next time. See you then, bye for now.