Greyfolk language’s monosyllabic roots and words: the background

Before I start talking about the nouns formed from the 20 monosyllabic roots in the greyfolk language, I want to explain some background concepts as well as the process. After almost two months, I finished these suckers about a week ago, and then I gave them a bit of time to rest because I knew that I would tweak them a bit more, which I did.

Hamming distance (which I have explained previously, and which I keep wanting to call hammerspace) was the key in determining which roots were usable. As previously discussed, roots that sound too similar aren’t ideal. So, I used Hamming distance to decide what “too similar” meant. In my case, it means that there needs to be a Hamming distance of 2 for things to not sound too similar. For example, «m» is a labial nasal and «n» is a coronal nasal, but there’s only one difference: the difference between labial and coronal. So, «m» and «n» have a Hamming distance of 1. However, «t» is a coronal plosive, so it has a Hamming distance of 2 from «m» (labial nasal), which is neither coronal or plosive. Yet, «t» only has a Hamming distance of 1 from «n» because they are both coronal. Thus, «tan» and «tam» are too similar but «mam» and «mat» aren’t. Furthermore, «nat» and «tan» are different enough because, even though «n» and «t» have a Hamming distance of 1, there are two instances of that difference, so that’s a total Hamming distance of 2 between those two words. It might seem tricky, but Hamming distance is easy to visualize.

Consonants Labial Coronal Dorsal Laryngeal
Nasal m n
Plosive p t k
Fricative s
Approximant j~ɰ1
Liquid l2
Transition h
  1. written «y», can be pronounced like English ‘y’ or ‘w’ or like Spanish soft ‘g’
  2. can be pronounced like English ‘r’ or like Spanish ‘r’ or ‘rr’

If any two consonants share a column or a row, they have a Hamming distance of 1. For example, «m» and «p» share a column, and «t» and «k» share a row. If they share neither a column or row, they have a Hamming distance of 2. For example, «n» and «y» (/j~ɰ/) are in different columns and rows. There is one big exception to this rule: «l» and «y» (/j~ɰ/) only have a Hamming distance of 1 even though they are in different rows and columns because many realizations of the liquid row sound like approximants.

Of course, I could also change the vowels and not just the consonants, but it’s not that easy. That’s because the first vowel dictates word class, which I also explained in the same post that I explained Hamming space. The “head-initial” vowels indicate words as follows:

  • «e» indicates a noun (or pronoun)
  • «i» indicates a modifier (e.g., adjectives, adverbs)
  • «o» indicates a verb
  • «u» indicates a function word (e.g., conjunctions, prepositions, particles)

And «a» is filler—it doesn’t mean anything except that the word isn’t over. So, it can’t be the first vowel.

Then, add the rules for syllables to start creating words. In the greyfolk language, the syllable structure is C1(C2)V(C3).

  • C1 can be «m n p t k s y l h»
  • C2 can be «y l», but not after «y l h»
  • V can be «a e i o u»
  • C3 can be «m n l»

A word just follows all of these rules. So, a word could be «me», «him», «pyo», «klul», «teka», «syepan», etc. Words can be written normally with spaces in between them, but this system has the advantage of being able to be written as a string of text with one minor adjustment. If a word—not a syllable!—does not have a C3, add a silent «h» to the end of the word. This disambiguates certain cases like «kamenyim» which would be «kamen» and «yim» or «kame» and «nyim». Using the silent «h» means that «kamenyim» is «kamen» and «yim» while «kamehnyim» is «kameh» and «nyim».

Now, I’ll return to discussing non-conflicting sounds. There is are two more rules to add to figure out Hamming distance between syllables and words in the greyfolk language. First, the difference between nothing and any sound is a Hamming distance of 1. For example, «nim» and «nyim» have a Hamming distance of 1 between them. «nim» does not have a C2 and «nyim» does, but they are otherwise the same, so this is a Hamming distance of 1. Second, the same root is allowed with different vowels. How else would it work? For example, «nem» and «nim» are fine because «nem» is a noun and «nim» is a modifier. Even if some vowels sound similar and get confused, because head-initial vowels determine word class, context makes up for the Hamming distance of 1.

Using all of these rules, there is a maximum number of non-conflicting syllables that can be formed, especially if they share a vowel. This was the hardest part of figuring out monosyllabic words. By hardest, I mean it was challenging and frustrating, and, yes, I did cry at least once. I have a very limited phonemic inventory, so there are a lot of constraints, and I put one extra constraint on myself: no monosyllabic words with a C2 and a C3.

What did I get?

This:

«me»
«se»
«ke»
«tle»
«yel» «yil»
«nel» «nil»
«ten» «tin»
«lem» «lim»
«pem» «pim» «pum»
«pli» «plu»
«min» «mun»
«kyu»
«kul»
«num»
«sul»
«lun»
«yum»
«myu»
«hu»
«syu»

With «nlu» left over.

So, that’s 20 monosyllabic roots to create 28 words. Not too shabby.

These words will be explained in following posts. I’m planning on discussing groups of roots. The other option is to go by word class, but that would be to show off the Hamming distance between each word in each class, but the above table can be used for that same effect. See for yourself!