Skip to main content
Loading Explore guides…
Phonology for Conlangers · Acedra
Home
Explore
Languages
Phonology for Conlangers
Sign in to save
Chat
If YouTube asks you to sign in, open the video on YouTube.
Watch on YouTube
Reload player
Transcript
74 segments
Copy
Hide video
0:00
Most languages are made up of sounds. That's what phonology is - the study of sounds, and if we want to make a realistic conlang, it's an area we need to understand well. There are all sorts of different sounds in the natural languages of the world, far beyond what we use in English. These speech sounds, or "phones" as we call them in linguistics, are created by changing how air flows through the mouth and nose. We can change the shape of
0:32
them by moving the lips, tongue and jaw and vary how the air flows using the larynx and lungs. Interesting though the details of this are, you don't need to learn a lot as a beginner conlanger. However, there are a few main points you need to consider. When discussing the sounds of speech, it helps us to be able to classify them into different groups. Probably the main such distinction we can make is between consonants and vowels. These are probably
1:05
terms you've heard before - at school or when learning a language, or indeed anywhere else, but whether you have or you haven't, let's define them so we're all on the same page. A consonant is a sound produced by restricting airflow in some way, while a vowel involves no such restriction, instead simply changing the shape of the vocal tract without closing it up too much. You may have heard that English has five vowels - A, E, I, O and U, but it's very important
1:41
to understand that these are *orthographic* vowels - that is, they're only the written letters used to denote vowels, while in phonology, we're more interested in the actual sounds we speak, regardless of how they're written. We'll go into more depth on writing in episode two, but it makes sense, right? English is infamous for spelling words the same that are pronounced totally differently, or indeed spelling two words differently when
2:11
they're pronounced the same. This means that the English alphabet is not a good way to write sound. Lets say you're making your conlang and you decide you want to add the /I/ sound, like in "lid", or "hiss". You decide to write this sound with the letter "I". Fairly easy, right? Well, maybe you have to leave your project for a few days and when you come back, you can't for the life of you
2:38
remember how that letter "I" was pronounced... was it /I/... or was it /ai/, or maybe even /i/!? And even worse, what if you give your work to your New Zealand friend, where the /I/ of "hiss" is pronounced more like /həs/! If only there were some sort of standardised way of writing sounds, so that no matter what dialect or even what language you speak, you can know 100% for certain how any sound should be pronounced. Well perhaps you've guessed, but actually there is.
3:14
This is the International Phonetic Alphabet, or "IPA" for short. It is exactly that - a different alphabet that's used by linguists and other linguistically-minded people to write sounds in a way that's completely neutral and universally understood. It's worth mentioning that, while that is it's aim, it does fail on those points at times - it was first created in the late 19th century and the essence of it has stayed largely unchanged since then, so it doesn't fully meet
3:49
the rigours of modern linguistics, but for most purposes, including ours, it solves more problems than it creates and is the standard taught to all linguists worldwide. It's entirely possible to listen to someone say something in their own language or dialect, transcribe it in IPA, then give it to someone else and have them replicate it almost perfectly. So, here is the official IPA chart. It contains every symbol and diacritic currently recognised by the International Phonetic
4:26
Association. It looks complex and it is a lot, but we can divide it down into simpler parts. For example, these three sections contain all the consonants, while this one contains the vowels. The three sections at the bottom then allow you to modify the basic consonant or vowel symbols to specify shades of difference that the plain symbols can't cater for. Let's start with the consonants. Consonants can be classified in a number of different ways, but the primary one
5:02
opted for by the IPA is by distinguishing where the air comes from. The table at the top contains "pulmonic egressive" consonants, meaning that sound is produced by pushing air out of the lungs. The second table contains sounds which are either non-pulmonic, meaning airflow is not created with the lungs, or ingressive, meaning air flows into the body, rather than out. For this video, we'll mostly ignore this table, although let's briefly look at the three columns it presents.
5:40
The first contains click consonants - rather rare sounds found only in one or two languages outside southern Africa. They're created by forming a vacuum between two points in the mouth, then suddenly releasing it at the front, allowing air to rush in. Examples are /ʘ/, /ǂ/ and /!/. The second category is implosives. These use a glottalic ingressive airstream, meaning air is made to flow inwards, before being pushed out by the lungs for the rest of
6:16
speech. They're found in around 13% of languages - particulary Central Africa and Southeast Asia. Finally, the table gives us ejectives, which use glottalic egressive airflow. They're very common in the Caucasus and North America and are created using a sudden burst of air from the glottis. So now let's look more at the pulmonic egressive consonants in the big table at the top. All of these we can describe in three aspects - voicing, place and manner. If you look at the table,
6:55
you'll see that every box contains - or at least has room for - two symbols. These are pairs that, while identical in almost every regard, differ in something called "voicing" - that is, whether the vocal cords are vibrating or not. Look at the box containing the letters "S" and "Z". These represent exactly the same sounds as the English letters make. Hold your hand against the front of your neck. Now, make a long /sssssssss/ sound. You may feel a few changes as
7:32
your throat naturally moves, but nothing much else. Now make a long /zzzzzzzzz/ sound. Feel that vibration? It's caused by two bits of muscle beating together over 100 times a second. You can try switching between the sounds without stopping the flow of air and you'll feel that your mouth and tongue stay exactly the same, it's just this voicing in your throat that differentiates the two sounds. Almost all sounds can be produced with or without vocal fold vibration. Where there are
8:09
separate symbols in the IPA chart for these two sounds, you'll find the voiceless one on the left and the voiced one on the right. Many sounds do have voiceless equivalents, but we only see one symbol in the box. These can be denoted using a special diacritic from the box at the bottom. The second way to describe a consonant is with its place of articulation. I've already mentioned how consonants are those sounds produced by restricting airflow. Since air travels all the
8:51
way through the mouth to get from the lungs to the outside world, we can interrupt it at quite a lot of places. It turns out that where you create the restriction changes the sound you produce. Place of articulation is listed along the top of the table, with each different place given a column. The further back in the mouth the restriction is formed, the further right in the table it
9:18
appears. Imagine a head looking to the left, with the columns of the table running through the mouth and down its throat. The lefternmost column is labelled "bilabial" which is fancy linguistics terminology for "using two lips". The cells listed below include sounds like /p/, /b/ and /m/. Take a moment to produce each of these sounds and feel how each involves you pushing both lips together.
9:50
If you try keeping your lips apart, you won't be able to make those same sounds, if any at all. Now let's look at the other columns. Remember we're working back, further into the mouth, so the next place listed is "labiodental", which means "using the lips and the teeth", usually the upper teeth and lower lip. You may recognise the English sounds /f/ and /v/. Next are the dental sounds, produced using the teeth and tongue. Most of the rest of the columns involve
10:27
the tongue, but don't use "linguo", meaning "tongue" in their name. English has the dental sounds /θ/ and /ð/ and some dialects also have others, transferred here from being "alveolar", which is the next place of articulation. Put your tongue tip on the roof of your mouth. Now run it forwards towards the teeth. Just before you hit the back of your teeth, you'll feel a little bump. This is called the alveolar ridge, and by placing your tongue against it, you can create probably
11:06
the largest group of consonants, the alveolars. English uses many of these - /t/, /d/, /n/, /s/, /z/, /r/ and /l/. American English also uses the tap /r/. Postalveolar sounds, as may be expected, are produced just behind this ridge and include the English sounds /∫/, /ʒ/, /t∫/ and /dʒ/. Retroflexes don't exist in most varieties of English, but are common in the languages of India, Scandinavia and Australia and are scattered across the world
11:49
elsewhere. They involve touching the tongue tip to the roof of the mouth, sometimes even bending it back on itself. Palatals involve the tongue body moving close to the very top of the mouth, an area called the "hard palate". English only uses the palatal /j/, but lots of sounds exist in other languages around the world. Velar consonants move another step back, employing the "velum", also known as the "soft palate" - the region at the very back of the mouth, just before you
12:27
reach the throat. English uses the velar /k/ and /g/ and, to varying extents, /ŋ/. Uvular consonants use that dangly thing behind the tongue as place of artiulation. French and German's "R" sound /ᴚ/ is uvular, as well as the Arabic /q/. Pharyngeals are formed at the pharynx, by pulling the very bottom "root" of the tongue into the gap within the throat. They exist in several Semitic languages, like Arabic and in some other languages elsewhere
13:06
in the world. The last column contains glottal consonants, where airflow is held at, or close to, the vocal folds themselves. The English "H" is an example. There are actually a few other places of articulation - mainly in between these main ones or created by using two other places at once. For example, the English /w/ is "labiovelar" - using the lips while simultaneously pressing the tongue
13:42
to the velum. These other sounds are found in the third section of consonants on the IPA chart. The last main part of any consonant is "manner of articulation". As we've seen, for any voicing and place of articulation, there can still be quite a lot of distinct sounds. In English, we have five voiced alveolars, for instance - /d/, /n/, /z/, /r/ and /l/. If you produce these, you'll be able to feel your tongue touching the alveolar ridge each time and you'll feel the vibration
14:22
of voicing, so what differentiates these? Well, a consonant involves a "restriction of airflow" - in this case using the tongue against the alveolar ridge. But what happens if we press the tongue against the ridge, or if we just rest it there? This will change how the air can escape - we can modify its direction, its speed, its pressure and what it hits along the way. If we press the tongue
14:54
firmly against the ridge and brace it against the teeth on either side, we can completely block the way out, stopping airflow for a short time before we release it. This is what we do to create the top row of sounds, called "stops" or "plosives". English has six of these - /p/, /b/, /t/, /d/, /k/ and /g/. All languages in the world have at least one, although the vast majority have at least three. The next row is "nasals", which also involve completely blocking the mouth, although
15:36
in these sounds, the velum, the part of the mouth used for velar consonants, drops down to allow air to flow out through the nose. Most languages have nasals, although a few lack them to some extent, most of which are spoken in the Americas. Trills and taps don't exist in most English varieties, although the alveolar ones are the short and long "R" sounds in Spanish and Italian. Trills are produced by allowing the airstream itself to vibrate an articulator, such as the tongue
16:15
or lips, while taps involve a single contact by consciously moving the muscle. The alveolar tap and trill are common in many world languages and the retroflex tap is found in some Indian, African and Australian languages, but these sounds are otherwise very rare, with some occuring in only a handful of languages. Fricatives are very common sounds, occurring in about 90% of languages. Most of those lacking them are from Australia or areas with very few consonants at
16:56
all. Rather than creating a full closure of air, like with stops or nasals, fricatives just narrow the passageway to being almost closed, so that air is forced through a very small hole and becomes turbulent. You'll see on the IPA chart that there are voiced and voiceless fricatives for every place of articulation in the main table, so take care to pay special attention to these sounds when
17:27
learning them, as several sound quite similar. The final manner of articulation is approximants. In these sounds, airflow is narrowed, but not as much as fricatives, so no turbulence is created. Instead, the sound is smooth. If you've ever heard the word "semi-vowel" before, these are a subclass of approximants, which are particularly similar to vowels. You'll see that there are actually two more rows on the table, containing "lateral fricatives" and "lateral approximants". All
18:07
the sounds we've looked at so far are known as "central". This means air flows over the tongue, which is often braced against the teeth at the sides. In a lateral sound, bracing occurs at the front of the tongue and air flows around one or both sides. Lateral approximants, like English /l/ or Italian /ʎ/ are most common, although lateral fricatives also exist, such as the voiceless alveolar /ɬ/ of Welsh and Greenlandic. One final manner of articulation isn't actually
18:48
shown on the chart. "Affricates" are sounds that begin as stops but end as fricatives. You may not have realised before, but the English sounds /t∫/ and /dʒ/ are essentially just combinations of an alveolar stop and a postalveolar fricative. These would be called "post-alveolar affricates". Other affricates exist, such as the alveolar /ts/ of Italian and German or German's labial /pf/. Note that not all stop-fricative sequences are affricates, though - only those which
19:23
behave like single consonants get the name. While we do have a /ts/ sound in English, such as at the end of the word "crickets", this is just a sequence of a /t/ and a /s/. While English words can begin with affricates, like in the words "chocolate" and "jelly", no native words begin with /ts/, so this clearly behaves like a cluster of two separate consonants. So, that's consonants explained. Before we go onto vowels, I'm going to take a moment to say that I
19:59
know this is all quite complicated, especially for beginners, but phonology and the IPA are things that you cannot conlang without - at least not to the extent you'd probably like to. This series should lighten up a little after this mammoth of an episode, but this is the groundwork that must be laid in order to progress. In future episodes, I'll take all this as more or less granted, so try
20:27
to spend the time before the next one's released trying to make sense of it in whatever way works best for you. I'll try to remember to chapter this video, so feel free to skip around as you need, to go back over anything you don't understand. As ever, I'll also include further reading in the description, including links to the IPA website and places you can listen to and type IPA symbols. So, onto vowels! There are a lot of parameters to describe consonants because they're usually
21:03
quite discrete. Although it does happen a lot, the phones generally don't float around and roam the mouth much. Vowels are kind of the opposite. By definition, vowels aren't produced by contact between articulators, so have far less to anchor them to any one point. They form more of a spectrum, which like everything in nature, we like chopping up into neat little boxes. The most clear distinction, however is in something called "rounding". Unlike consonants, vowels are
21:40
mostly produced in the oral cavity, using only the tongue. However, this means there are parts of the mouth that aren't doing anything, as they would be in consonants - namely the lips, teeth, nose and throat. These can all be used to modify the basic vowels, but by far the most universal is by employing the lips. We call vowels that involve protrusion or shaping of the lips "round vowels",
22:10
because the lips generally take on a circularish shape. The vowels that don't do this are termed "unround" vowels. On the vowel quadrilateral, which you'll find just under the consonant table on the right-hand side, rounded and unrounded vowels are shown next to other, like voiced and unvoiced consonants are, with the rounded vowel on the right and the unrounded vowel on the left. It's important to note that many of the contrasts made in consonants are not
22:44
present or rare in vowels. Vowels are almost always voiced, for instance. Voiceless vowels exist in a few languages, like Japanese and some American languages, but are very rare and usually not important to the languages they do occur in. Likewise, since vowels don't restrict airflow like consonants, place and manner of articulation are not helpful descriptors for them. Instead, we describe the shape and position of the tongue in the mouth, using the terms "height" and
23:22
"backness". "Height" refers to how close to the roof of the mouth the highest point of the tongue is. A high vowel, also called a "close" vowel is one where the tongue is raised in the mouth, while a "low" or "open" vowel is one where the tongue sits very low in the mouth, leaving a big space above. Vowels sitting between these two extremes are called "mid" vowels. Some of the high and high-mid vowels in English are /i:/, /u:/, /ı/ and /ŭ/, while some of the low and low-mid vowels
24:02
are /a:/, /ä:/, /ɛ:/, /ɔ:/ and /ò/. "Backness", as you may have guessed, refers to how close to the front of the mouth or to the throat a vowel is pronounced. Front vowels include /i:/, /ɛ:/ and /a:/, while back vowels include /ʌ:/, /ɔ:/ and /ò/. We don't have many central vowels in English, although many dialects, including my own, contain a handful, like /ʉ:/ and /ä:/. The one found in almost all dialects of English, though, is the central mid unrounded vowel - the vowel that
24:45
lies right at the centre of the quadrilateral. This vowel has a special name - schwa. It's very common in Germanic languages, like English, as well as a number of others worldwide. In addition to single vowels, languages can also combine two vowels into one sound. We call these "diphthongs". This is a concept a lot like consonantal affricates, where the two vowels are treated as one entity. Most languages have diphthongs,
25:23
including English, such as our /ai/, /au/, /ei/ and /əu/. Just like with affricates, it's totally possible to have two adjecent vowels without them becoming diphthongs. We describe these vowels as being in "hiatus". In fact, the word "hiatus" contains vowels in hiatus. It's composed of three syllables - the first ending in a diphthong and second beginning with one. It's also possible to have individual vowels in hiatus, such as in the word "museum".
26:04
Just like consonants, there are a handful of adjustments we can make to our mouths to create vowels not shown on the chart, such as letting air flow out of the nose as well as the mouth, which we call "nasalisation" or changing how long we hold the vowel for, which we call "length". One I will touch on, though, is tone. In some languages, it doesn't just matter how a vowel is formed, but also at what pitch it's spoken. A particularly famous example is Mandarin, where a
26:42
word changes meaning depending on whether it's produced with a level high tone, a rising tone, a falling tone or a falling-then-rising tone. Tone is incredibly common, occuring in over 40% of languages, but concentrated in Africa, East Asia, America and the Western Pacific. Most tonal languages only distinguish two or three tones, while the rest use more complex systems. So that's the basics of the sounds themselves. Now let's see how languages arrange them. In English,
27:25
we usually mark a noun as plural (that is, show it as representing more than one individual) by adding an "S". So "cake" becomes "cakes", "raven" becomes "ravens" and "potato" becomes "potatoes". In each of these words, we add an "S", but let's take a minute to listen to how we actually pronounce this sound. In "cakesssss", we hear a voiceless alveolar fricative - an "S", as expected. But in "ravenssssss" and "potatoesssssss", do you hear that? It isn't
28:04
actually a voiceless sound, but rather a voiced one. So what determines whether we use the voiced or voiceless sound? Well, in "cake", the final sound is voiceless, while in "raven" and "potato" it's voiced - being an approximant and a vowel. So after voiceless phones, we use a voiceless version of the sound, while after voiced phones, we use a voiced version. These two different versions of the sound are called "allophones". Allophones will always be completely predictable
28:43
from their context. The voiced stops in Spanish vary allophonically. Between vowels or in some other environments, they're pronounced as voiced fricatives of the equivalent place of artiulation, while elsewhere, they remain as stops. Since the separate allophones of a certain sound never occur in the same context, we refer to them as being in "complementary distribution". The "phonemic inventory" of a language is the set of meaningful sounds it contains. This is
29:19
generally what you'll create for your conlang. While allophones are important in a language, we generally don't include all of these in our inventories. This is because phonemic inventories only contain a language's "phonemes". A phoneme is a meaningful unit of sound. In our Spanish example, /d/ and /ð/ are allophones. This means using one rather than the other might make you sound a bit weird, but it won't change the meaning of what you're
29:53
saying - they're two realisations of the same phoneme, so only one entry is made into the inventory and we can just note the variation elsewhere, such as in a "phonetic inventory", which lists all the sounds of a language regardless of whether they're meaningful or not. A "syllable" is a unit of speech made of a number of phones. It generally relates to how intonation is placed within a word and is made up of one vowel or diphthong and its surrounding
30:28
consonants. The vowel portion is called the "nucleus", the consonants before it the "onset" and the consonants after it the "coda". Different languages allow different types and sequences of consonants in the onset and coda. For instance, Kirundi allows any of its around 25 consonants in the onset and any of its 5 vowels in the nucleus,
30:58
but never uses a coda. English, on the other hand, allows up to three consonants in the syllable onset and five in the coda, although only allows certain sequences in both positions. The smallest syllable in English contains only a vowel, meaning that the onset and coda sections are both optional. We can transcribe syllable structure using capital letters as variables - a "C" stands for any consonant, while a "V" stands for any vowel. Brackets show optional segments and
31:38
if an onset or coda sequence gets too long, we can simplify it using superscript numbers. Kirundi's syllable structure can therefore be represented like this, while English's can be written like this. The way sounds are built together into valid syllables is termed "phonotactics". Alright, so that was a lot of dense linguistic information. Now let's do some conlanging. When I start putting together the sound inventory for a conlang, there are a few
32:15
sounds I always add. These are the most common ones, such as the bilabial and alveolar nasals, bilabial, alveolar and velar voiceless stops, the voiceless alveolar fricative and at least one of /w/, /j/, /l/ and /r/. Of course, not all of these exist in every language - in fact, I'm not aware of any sound that exists in all languages, except in the most generic terms - but this is a good
32:48
minimalistic starting point we can build up from and there's always the option to remove some if we see fit. If you're making a non-naturalistic conlang, you can also do whatever you want, because you're not bound by realism. To explain my decisions in this first step - the bilabial and alveolar nasals are the most common nasals, with the bilabial actually being the most common phone overall. Voiceless stops are more common than their voiced counterparts and most languages
33:23
have consonants from at least the three places of articulation given. The alveolar fricative is also very common and voiceless fricatives are more common than voiced ones, though not by as much as in stops. Some sort of approximant is also very common. /j/ and /w/ are each found in over 80% of languages, so adding these is a good idea, while /l/ and /r/ are also very common. In terms of vowels, all languages have at least two. These will always contrast in height,
34:02
so a language cannot have only /i/ and /u/. Very very few languages have just two, though. Three-vowel systems are quite rare, but exist in a good few languages and five-vowel systems are very common. Four-, six- and seven-vowel systems are also pretty common. /i/ is the most common vowel, but all languages have some sort of low vowel. For now, let's stick with a basic three-vowel setup of /i/, /u/ and /a/.
34:36
The next step I'd take is to think about what I want my final language to sound like - its "phonaesthetics". If I want to imitate a natural language, I'd need to choose sounds from that language. Wikipedia's quite good for this. If you look up a language you're interested in, it usually provides phonemic inventories of both the consonants and vowels and gives other important information, like tone, prosody and syllable structure. Let's say I want my
35:10
language to sound like Arabic. Arabic has lots of fricatives, gutteral sounds, such as velars, uvulars, pharyngeals and glottals, a series of what are called "emphatic consonants" and a missing voiceless bilabial plosive, which is instead replaced by a fricative. Personally, I can't pronounce the emphatics, and I'd like to be able to speak my conlang, so I'll remove them. I also don't like the postalveolar fricative and affricate pair, so I'll tidy
35:44
that up a bit and make them both fricatives. Just for fun, I'll also add the velar nasal to complete the trio. Arabic has a three vowel system, but also distinguishes vowel length. I'll keep this, but also add in a schwa, which won't distinguish length and can only be short and a high central vowel which can only be long to complement it. Arabic syllable structure is quite simple. Let's change it slightly so that a second consonant is allowed in the onset,
36:19
but only if it's a semivowel or "R". Likewise, the first consonant of the coda must be one of the four liquids or two nasals. Finally, a little allophony. Let's say that vowels nasalise before nasal consonants - that is, they are pronounced with an open velum, like in French "chance". A nasal consonant is homorganic before a stop - meaning it takes on the same place of articulation and voiceless stops become glottal when at the end of a word.
36:54
If you don't want to copy a natural language, you can also do it yourself. To build up an inventory from scratch, it's helpful to realise that there are trends and patterns in sounds. As mentioned, some sound types are more common than others, but phonemic inventories also have a certain symmetry to them. Let's say our language will distinguish bilabial, alveolar, palatal, retroflex and velar stops, like in many Australian languages. Now let's say I
37:28
want to add voiced and ejective alveolar stops. We generally wouldn't see only these added, unless there's a good historical reason, but we'd see the full "series" added, giving us voiced and ejective stops for most or all of these places of articulation. There can of course still be "holes" in the system, where we maybe don't have the bilabial ejective or the voiced palatal, but this then becomes the exception, rather than the rule. It's most common that we see this sort
38:03
of symmetry in stops, but it can also occur in other classes, if there are enough contrasts in voicing or another feature, and we can also see symmetry between classes - perhaps our language will also have the five nasals corresponding to our stop series. It can also be fun to add some random sounds to an inventory. A lot of sounds in the IPA are quite rare, but that doesn't mean we can't use them. If we do, sound symmetry might be a good way to squeeze them in. Rather than adding
38:38
a lone uvular nasal, we could add a uvular stop and maybe a fricative to complement it. Otherwise, we could just add it on its own, so long as there aren't too many random things. Or, it could be an allophone of another sound. Even if you're not mimicking a natlang, I recommend you read as much as you can on the phonologies of different languages. After a while, you'll start to get a sense of how they tend to work and can then apply this to
39:08
your conlangs. If you're wondering how to go about creating realistic allophonic variation, reading up is again a good idea. You can extract some of the ideas I've used in this video, but the basic rule is that allophones are generally quite similar to their mother sound. Just shift a voicing or place of articulation a little, but remember how each sound is produced. Bilabials aren't going to suddenly become retroflex, but they might shift to labiodental or even
39:41
labialised velar, where the lips are still used. Phones may also be affected by those around them, becoming more similar to a sound directly before or after. The basic patterns of allophony are the same as longer-term language evolution, so we'll talk more about it in episode 10. So, just a couple more things to mention before we reach the end of the video. So far, we've only talked about spoken human languages, but this isn't the only option. Firstly, only around 95%
40:17
of the world's languages are spoken. There are thought to be about 300 sign languages in use, and it may surprise you, but these actually have phonologies as well! In its strictest sense, "phonology" refers to sound systems, but obviously sign languages don't have any sounds. Instead, we apply the term to the similar concept they do use. In spoken language, there's a system of sounds specific to that language. Likewise, sign languages each have different ways that
40:53
signs can be formed. Rather than voicing, manner, place, rounding or any other parameters of spoken phonologies, sign phonolgies use features such as handshape, location of signing, facial expression, movement type and orientation of the hand during the sign. In some sign languages, there are handshapes or movements not used in other languages, while within a language, you can make a
41:24
slight change to one of the parameters and change the meaning of the word, just like how changing a voiceless sound to a voiced one distinguishes the words "feel" and "veal" in English. If you want to make a signed conlang, these are the kinds of things you'll need to understand and study. Our basic understanding of phonology can also be changed if we want to make a conlang for an animal or alien race. The sounds we can make and understand are governed by our anatomy. In a bird,
42:00
for instance, it wouldn't make much sense to include labial, dental or alveolar sounds, because they require the lips and teeth, which birds don't have. A bird conlang would probably put a much higher emphasis on suprasegmental elements, like tone, length, stress and rhythm, as in nature, birds communicate through their songs, so any language that evolved in birds would likely make use of the same call-producing mechanisms,
42:31
or at least be governed by the same anatomy. For any non-human conlang, you'd have to research the animal you're interested in or that your potential alien race may have similarities to, to understand what their limits and capabilities are. Non-human conlanging isn't an area I'm very experienced in, so if you want to know more, I've left a very good video link in the description.
42:58
And with that, we come to the end of the video. This one was a mammoth to make. It's by far the longest video I've ever made, but it's very important to me that this series is as complete as possible. That said, the next episodes should be much shorter, as any major topics I've already split up in my planning. I'm not sure when the next episode will come out, as the plan I laid out in the introduction has immediately gone out of the window,
43:31
but I'll try to get on it as soon as possible. If you're following this series, let me know in the comments and feel free to ask any questions - I'm sure you'll have some. If you missed the introduction video or are watching in the future, when there are already more episodes out, check out the full series playlist on screen. So, until next time, thank you for watching!
Here to help you learn
Sign in to keep a private conversation with your course tutor.
Ask me anything about the material…
Sign in to chat