All language is encoded in fundamental repeating units to convey meaning. For a spoken language,
those units are sounds. This is the International Phonetic alphabet, or IPA. It's is a collection
of every sound from every known language on earth. No language could possibly have all
of these sounds, or even most of these sounds, but there's no sound the human vocal tract
can produce consistently that isn't featured somewhere on this chart. Every language will
have what's called a sound inventory, or sound chart that displays exactly what
sounds are present, all of which are written in the standardized IPA notation. For your
language, the very first thing you'll want to do is make a sound chart of your own. Have
a look through the IPA and see which sounds you like. Very conveniently, Wikipedia provides
audio files for most of these sounds, so you can hear exactly what they're like before
you include them. If you want your language to sound like any real-world language in particular,
you can also look up the language and see its own chart. Be aware that the symbols that
the IPA uses to represent a sound may not be the same as the one English uses. For example,
even though this sound is represented with the letter j, it actually stands for the sound
/y/, whereas this, this, and this are actually /j/, /ch/, and /sh/. Some sounds are much
rarer than others. For example, as English speakers we're familiar with the sounds
/th/ as in 'think', and /dh/, as in 'this', but it turns out these are both pretty rare
cross linguistically. The uvular stop /q/, found in Arabic, Mayan, and some Inuit languages,
is also quite rare, and the lateral affricate /tl/, found in Nahuatl and Navajo, is even
rarer than that. If you're making a fictional universe featuring many languages, it's
good to be aware of these and keep the relative cross-linguistic frequencies of these sounds
in mind. If you're having trouble thinking of what sounds you'd like to include, try
coining some sample words that you like the sound of and listing the sounds that they
contain. But it's not as simple as grabbing any random sounds that tickle our fancy; we're
dealing with a naturalistic conlang here, and all natural languages share some features
in common, and in terms of phonology, one very strong tendency is sound symmetry. In
the IPA, the columns indicate where in the mouth the sound is produced, starting at the
very front with both the lips, and moving back towards the glottis. The rows represent
the way the sound is produced, whether it's a stop, produced by completely obstructing
airflow, a fricative, produced by leaving a small gap for air to hiss or buzz through,
an affricate, produced by initially obstructing airflow in the same manner as a stop before
releasing in the same way as a fricative, a nasal, produced by letting some air pass
through the nose, or a liquid, produced by slightly changing the contour of the unrestricted
airway. Any natural language is likely to have more than one sound in any given row
or column. There may be one or two exceptions, but for the most part, the sounds present
in a language should come in series. For example, we would never expect to see a natural language
that features these sounds INSERT THING HERE Whereas something like this is much more likely
to occur naturally. Now, notice how all of these spaces on the chart contain two symbols.
The ones on the right are voiced, meaning they require the use of the vocal cords, while
the ones on the left are voiceless. All of these sounds are collectively called obstruents.
All natural languages in the world feature at least some voiceless obstruents, but not
all natural languages have voiced obstruents. If a language has a voiced version of an obstruent,
then it will very likely have its voiceless equivalent. These sounds are called resonants
or sonorants. All languages in the world will feature at least some voiced resonants, but
voiceless resonants are exceedingly rare. Among consonants, resonants include nasals
like [m] and [n] and liquids like [r] and [l]. Any natural language is very likely to
include at least one nasal, in fact I have yet to come across any natural language that
doesn't have a [n] sound, and one liquid. Vowels are also counted as resonants. Every
language in the world features at least two vowels. Some like Arabic and Yupic inuit have
only three vowels, some Native American languages have four vowels, but by far the most common
layout is the classic five vowel system, which is common in every language family on every
continent. Some languages like Turkish have as many as eight vowels, and some languages
go even further than that; as much as we'd like to think that English has five vowels,
it actually has something more like 20. How many sounds your language has is entirely
up to you. Some languages like Hawaiian, Rotokas, and Mohawk have less than a dozen consonants
in their entire phonology, while others like Ubykh and Taa have over 100. Most languages
will have somewhere between 20-30. It's always best to err on the side of too few
sounds, as a smaller inventory will be easier to work with and help give your language a
more distinct and recognizable phonetic character. Plus, there'll be plenty of opportunity
to add or remove sounds as we evolve the language further. From these, we can compile all the
sounds in the language into a single chart. So for our sample language, we'll start
out with the basic sounds: /p/, /t/, and /k/ are extraordinarily common, as are [m] and
[n]. For now, we won't include any voiced sounds, but that can be remedied later. As
for fricatives, /s/ is extremely common, so we're safe including that, and the glottal
fricative /h/ is fairly common too. But just for fun, let's throw in something a little
less common like the lateral fricative /hl/. This is a pretty rare sound, but can be found
in welsh, zulu, and Navajo. For affricates, let's throw in a /ts/, which is a fairly
common affricate and matches the other alveolar sounds we've got. /ch/ is also extremely
common, but I'm choosing to omit it for simplicity, and I'll need it to help illustrate
something later on. Keeping the idea of sound symmetry in mind, we also might as well throw
in the lateral affricate /tl/ to match the lateral fricative. Again, we're not using
voiced obstruents at the moment, so we won't include /z/ or /dz/, or /zhla/ or /dla/.
There are many different variations of the r-like sounds, from /r/ to /r/ to /r/ or even
/r/, but the most common is the simple tapped /r/, so we'll use that one. L also has a
bunch of variations, /l/, /L/, /ly/, but we'll stick with the basic /l/. Finally, lets throw
in /w/ and /y/, both very common sounds, and voila, we have our consonant chart.
As for vowels, we'll just stick with the basic 5-vowel system, since it's the most
common. I'm also going to throw in vowel length as a contrastive feature. A long vowel
(usually) sounds more or less the same as its short equivalent, just held about twice
as long. Having a distinction between long and short vowels is extremely common across
the world; English kind of has long vowels, but they only appear in specific phonetic
environments. Compare the pronunciation of back vs. bag. However, languages like Latin,
Finnish, or Mongolian pay much closer attention to whether a vowel is long or short. Long
vowels can also be very useful later on, so we'll include a long version of every one
of our five vowels. So there we have it, here's our finalized
sound chart. It's pretty basic, but it'll serve us well when we make the proto-language.
But as you'll see, there's going to be plenty of opportunity for modification later.
Now for a quick note on Romanization: you may have some IPA symbols that aren't practical
to use when transcribing your language, so you'll have to choose easier characters
to represent these sounds. It is strongly advisable to choose symbols to represent these
sounds that are as easy as possible to interpret. For our language, let's represent the /y/
sound with a y rather than a /j/ most english speakers would mistake that for a /j/. Other
than that, The only sounds in our language that are a bit unwieldy to write in IPA are
the lateral fricative and affricate. Most readers would have no idea what sound to make
for these symbols, and it would be a real pain to have to look up and copy and paste
these symbols all the time. Instead, lets simply represent them with a /hl/ and /tl/.
These are what we call digraphs, where a single sound is represented with multiple characters.
Now, these might be misconstrued as consonant clusters, so someone might be tempted to read
/atla/ as /at-la/ instead of /a-tla/, but as long as we detail this in our romanization
system, and have a way of potentially disambiguating any similar consonant clusters that occur,
we should be fine. And then there's the issue of representing long vowels. A frequent
way to do this is by simply writing the vowel character twice to indicate that you hold
it for twice the length. That makes enough sense, but unfortunately, English's weird
spelling system means that more often than not, an English speaker will interpret "ee"
not as "e:", but as "i:" and "oo" not as "o:", but as "u:". There are
a couple of ways around this, but I'm going to take the simple way out and just use vowels
with macrons, just as latin does. But we're not out of the woods just yet: a languages
sound system is more than just a collection of sounds. Every language has a series of
rules that govern how sounds can be put together. This is part of the reason langauges have
recognizable phonetic qualities; we can often make a pretty good guess at what language
a person is speaking even if we don't speak it ourselves.
So, we need to set some phonotactic guidelines. The first and most basic rule to set is syllable
shape. Any given syllable consists of up to three segments: the onset, the consonant that
begins the syllable, the nucleus, the part of the syllable capable of receiving stress,
which is almost always a vowel, and the coda, the consonant that ends the syllable, if there
is one. Keep in mind, for a consonant to qualify as a coda, it needs to precede the onset of
another syllable or appear at the end of the word. If a syllable has no coda, it's called
an open syllable, and if it does have a coda, it's a closed syllable. About one eighth
of the words languages don't allow codas at all: Hawaiian, Swahili, and Guarani, consist
of only open syllables. Some languages like Siglitun and many Asian languages allow only
a very small number of consonants to serve as a coda.
Most European languages are considered highly complex in terms of codas, like English's
notoriously dreadful "strengths" and "angsts". And some languages can go even further than
that: It is possible, if not likely, that your syllable shape will change as you evolve
the language. To illustrate this, and to keep it simple, I'm going to say that our proto-language
is going to consist of purely open syllables. Another rule you'll want to establish early
on is how the stress system works. In most langauges, a certain syllable of any given
word will receive the stress. Some languages have a fixed stress system, that is the stress
always falls on the same syllable no matter what. In other languages, changing the stress
can change the meaning of the word, compare the English present and present, and in yet
others, which syllable is stressed depends on the phonetic environment. In a language
like latin, the third-to-last syllable is stressed UNLESS the second to last syllable
contains a long vowel or has a coda, in which case, it receives the stress. Since our sample
proto-langauge is purely CV, we don't have to worry about coda consonants. Let's say
that the stress always falls on the third-to-last syllable unless the last-to-last syllable
contains a long vowel. And there we have it. One proto-phonology. We now have everything
we need to start coining words. Join me next time when we'll start figuring out how to
put words together.
For more infomation >> How to financially prepare for a disaster - Duration: 2:41. 
For more infomation >> Scalable Business: How To Run a Business -- NOT a Hobby - Duration: 2:54. 
For more infomation >> Ford focus saat tarih ayarlama nasıl yapılır? (How to set the time and date?) - Duration: 1:10. 
For more infomation >> 2-How To Insert Data in Database Using C# with SQL - Duration: 6:17. 
For more infomation >> Tracking down a fugitive: retired U.S. marshal explains how it works - Duration: 2:27. 
Không có nhận xét nào:
Đăng nhận xét