Previously: Conlanging I
I'm making a language. The last post on this blog was a log of my design thoughts, but it got quite large, so here's part two. I'm updating it semi-daily. I'll post a finished reference grammar for the language around the end of April.
Yesterday, in the last post, I said that I wanted to start making noun classes for the language, and I wanted to start with perdurants (things like events, states, and processes). I started by looking at words X for which you can say "it happened during the X". I don't think that these nouns with duration are the same as perdurants; my first guess is that they're just events and processes. This category of nouns with duration includes words like: wedding, launch, haircut, vigil, rampage, assault, voyage, turbulence, lottery, fiasco, improvement, and ovation. Lots of the words in this category have senses that are mass-nouns, like "some drudgery", "some improvement", "some depreciation", and "some contact". I'm going to call count-nouns that have duration "events", and mass-nouns that have duration "processes". Event nouns are often made from verbs by zero-derivation, meaning that you keep the word as it is with no new affixes or ablaut or anything. The word "run" in the infinite verbal construction "to run" becomes, by zero-derivation, an event, "it happened during the run". Verbs with perfective aspect become events and verbs with continuous aspect become processes, for example, the word "running" in the continuous-aspect verbal construction "he is-running" becomes a process, "some running", by zero-derivation. Not all nouns with duration fit this pattern, like who ever heard of "vigil-ing"? That's right, Urban Dictionary has heard of it, but it's barely an English word if it is at all. "Drudge" is an archaic verb, but the noun form referred to a person who did drudgery, and not to the event of "a drudging".
Anyway, I don't think the Xenants will have both event nouns and process nouns. Maybe they'll have something like "manner nouns" and aspect will be encoded on the verb "exists". I'll have to do some experiments and see how that looks and works, particularly with relative clauses. Eventually I also want to figure out some fine-grained semantic categories for the events/processes/manner nouns.
If those are events and processes, then what are states? If you can say of a word X that something is "in a state of X", then I will take X to be a state. Lots of the words that fit in for X are nominalized adjectives. The adjectives (sensitive, tranquil, lawless, dizzy, quiescent, turbulent, ignorant) become the states (sensitivity, tranquility, lawlessness. dizziness, quiescence, turbulence, ignorance) for example. In Conlanging Part I, I said that when you nominalize an adjective, you get a dimension/quality/property. But here we're getting states by nominalizing adjectives. So are dimensions the same as states? No; "Activity" as a dimension is space of possible values. Being in "a state of activity" means that something's existence can be (continually/indefinitely) described as having a high value within that space. Any dimension, any nominalized adjective, can be treated as a state in that way - there is systematic polysemy - but we shouldn't forget the distinction between the senses, the distinction of (abstract descriptive scales) versus (existence of entities being described based on their values on those scale).
What about other states that aren't nominalized adjectives? Some states are made by nominalizing verbs. I'm not sure if there are any restrictions on what sorts of verbs can be nominalized to produce states. One state-producing affix for verbs is -ment, as in (abatement betterment commencement derangement embarrassment fulfillment government harassment impairment judgement management nourishment payment recruitment settlement). Another derivational affix is -ion on verbs that end with "-ate" (why are there so many?), giving states such as (ablation calibration dedication elevation fixation generation hallucination illumination lactation manipulation negotiation operation penetration radiation saturation termination ventilation), or the affix -ation on some other verbs that don't end in "-ate", (adaptation, cancellation, documentation, examination, fermentation, imagination, limitation, permutation, reformation, transformation, variation). And some other verbs that don't end in -ate are nominalized to states with the suffix -cation, giving states like (application, classification, identification, modification, specification, verification). I bet that's something I'd understand if I knew more Latin. Oh well. A few verbs that end in "ict" can be nominalized to states by adding "-ion", such as (addict, afflict, constrict, contradict, depict, evict, restrict). Also, "derelict" has senses as a noun or an adjective, but for some reason we still have "dereliction". I think "-sion" has a genetic relationship with "-tion" through historical sound change, and states ending in -sion include (confusion, depression, suspension, tension) which are probably derived from the verbs (confuse, depress, suspend, tense), or the past participle forms of those verbs, which can function as adjectives (confused, depressed, suspended, tensed). And maybe all states that seem to derive from verbs are actually doing so by means of a past-participle adjective in the middle of the derivation path, in that same manner? I don't know.
There are a bunch of states that end in -tude, like "disquietude" and "pulchritude" but most of them don't see much common use. I'm also not sure what the pattern is for their derivation. A few seem to come from verbs ending in "ify", like (amplitude, beatitude, certitude, fortitude, gratitude, magnitude, rectitude). We also have nominalizations of all of those verb using the suffix -ation (amplification, beatification, certification, fortification, ...). A state of rectitude versus of state of rectification? Is there a difference or are the affixes redundant? I think rectification suggests a change to a state of rectitude, maybe? It's a really small difference if anything. Let's move on. Some other -tude words are plain adjectives with "itude" added on, like (aptitude, exactitude, ineptitude, inexactitude, longitude, promptitude). A few other -tude words are less clear to me in their derivation, but they're such rare and clunky words, I'm not even going to bother trying to figure out their sources of derivation by looking up the etymologies.
Some other state words end in -ance or ence. I'll analyze those tomorrow. There are a few verbs ending in "-ing" that should probably be thought of as processes, but it's tempting to call them states, like "being, "becoming", and "waiting". Why isn't there a word for the state of waitment or a waititude or a waitation? I need to think more about the relationship between processes and states.
In the list of states that I started generating to get a feel for what types of states exist or are worth lexicalizing, most of the remaining ones were epistemic states or emotions or other biological conditions (acidosis, agony, alarm, anger, awe, bliss, decay, delight, denial, despair, disbelief, euphoria, frenzy, health, joy, ketosis, melancholy, nirvana, pain, panic, peace, repair, repose, rest, reverie, shock, sleep, slumber, turmoil, wonder). I could pad that with more of the same, but you get the picture.
Here are the remaining states that I haven't categorized yet: (chaos, disarray, disequilibrium, disrepair, equilibrium, flux, grace, limbo, neglect, purgatory, repair, ruin, stasis, undress). Some of those are also verbs. Maybe it's a zero-derivation thing. Not all though. Tricky.
That's it for tonight. Soon I'll pick some more normal word classes for the Xenants' language like "material substance" and "organism" and "role for decision maker". That should go faster and require less thought. Sounds nice, right? Absolutely nice. Goodnight.
-
Hello again! To identify events and processes, I asked whether you can speak of things happening during them, and I thought that perhaps you couldn't do that with states. But now I'm less sure of that. You can say, "it happened during the (tranquility, depression, frenzy, tension, hallucination)" and you can also speak of entities existing in those states. With states, it often makes more sense to say "it happened during $a-specific-thing's (tranquility, depression, frenzy, ...), but I don't think it's necessary. So what's the real difference between states and other perdurants? It's easy enough to say that states are uniform whereas processes have distinct phases, but that doesn't help me to categorize words as much as a lexical/phrasal/sentential test would wherein I could check a word for appropriateness of use.
What about massed usage of state nouns? You can talk about having "some tranquility" or being in "a frenzy", and you can do either with "depression" (have some or be in a). Does that reveal something about the semantics of the words? Maybe.
Maybe a state is always relative to a specific object's existence, whereas an event or process can have participants, but not one specific object that they depend on. But that thing can be a very general thing, like "all of reality around us". Mmm, okay, states might have a special dependence on things that events and processes don't also have on their participants, but even if so, I don't want to define statehood for the Xenants in terms of an intuitive notion of dependence any more than an intuitive notion of temporal uniformity.
I did a lot of work on clustering events/processes into semantic classes. It would be nice if I could forget about states for a second and show you that work, but it's still messy and not ready to publish.
New hypothesis! Lots of event-words and process-words come from verbs by zero-derivation. States, I contend, also come from verbs, but only a special class of verbs. Consider: when you say "the dog is brown", the syntax might suggest to you that "brown" is an verbal argument, but semantically, it makes more sense to consider "is brown" as a phrasal verb. That's the logical predicate. State words, I contend, derive cognitively (if not lexically) from copular phrasal verbs like "is-brown". To be in a state of chaos is to "exist-as-a-chaotic-thing", and that should be thought of as a single verb. In support of this conception of states, note that "state of chaos" is synonymous with "state of being-chaotic". Also, a bunch of the states looked like nominalized adjectives, just like "chaotic thing" is a nominalized adjective. And other states looked like nominalized verbs, but past participle verbs act like adjectives, so maybe those state-words are also secretly nominalized adjectives.
How does this copular conception of states interact with the aspectual hypothesis of the distinction between events and processes? Inceptive and terminative aspects on copular phrasal verbs seem to encode stative events ("it happened during the event of it beginning to be a chaotic thing"), and continuous aspect on a copular phrasal verb seems to encode a process ("we did some existing as chaotic things", "it was in a process of existing as a chaotic thing"). Putting the perfective aspect on copular phrasal verbs feels weird though. Maybe there are perfective stative events, maybe there aren't. I'm having some trouble thinking of them.
Based on the above, I think the Xenants won't have state words. They'll instead have adjectival nouns like "chaotic thing". And as for non-copular events and processes, I think I should finish organizing them into semantic classes before I commit to replacing them with (aspect-agnostic) manner nouns. Some events have location as well as duration ("we went to the funeral", "we did it during the funeral"), and it feels a little wrong to treat locative events as verbal manners, like in "we caused his mind to be lost to future generations by means of a funeral". You obviously can use locative events as manners, but I don't want to restrict them to being only manner-nouns if they can productively pair with other instrumental affixes like the locative/directive frames. Maybe it would help me if I made a non-English distinction between funerals as plan descriptions and funerals as plan executions? Hm.
Okay, for nominal classes, I want a class something like (event or manner). Also something like (dimension or adjectival noun), maybe both, though that feels redundant. I want at least one class for physical materials and one for physical objects. Maybe there will be a class for materials by microscopic composition and arrangement (cotton blood diamond glass), and one for macroscopic forms of materials (foam fluff powder tar mist) (and maybe thermodynamic material phases like (liquid) would go in the same category), and one category for functional material roles (food drug fuel adhesive lubricant surfactant irritant). There might even be classes for specific material roles, like (tofu wine feta mayo cake) could be in a class just for materials categorized by their function as food, although objects can also function as food, and there's often systematic polysemy in English of food words having object-senses and material-senses, like ("a cake, some cake"), so that deserves some consideration. Will bare food nouns only be a substance for them, and countable food item will require a measure phrase, like "a loaf of cake" or "an each of cake"? Or maybe there will be a word class that has systematic polysemy. Or something else. I could also categorize physical materials by origin, like "bodily substances" or "diagenetic rocks".
It seems ontologically right and basic to have noun classes for time periods and spatial regions, but I'm going to hold off on those. I do want some categories for abstract objects, but I need to review my past ontological work to get a good perspective on exactly what abstract object classes are good at dividing up the space. The set of (linguistic objects, mental objects, social objects) are a common trio of abstract classes in formal ontology, but I'm not just going to import that without thinking about it.
That's it for today. Goodnight, pumpkin butt.
-
Good morning, reader. When analyzing immaterial entities into semantic clusters, I at first found myself making some categories like "legal entity", "financial entity", "political entity", "mathematical entity", "software entity", "musical entity". I don't want the Xenants to have ontological classes like those. That isn't a regular systematic way of categorizing objects or masses. There is no limit to how many Xs you can use to make categories of the form "X-pertaining entity". So I need some other organizing principle than "domain categories". Preferably a few binary criteria. Also, what do you do when something falls into more than one domain, like how a ballad can be a poem or a song? There's probably another solution, but my solution is "no domains".
We could categorize symbolic expressions by roles like "description" and "designation". We could distinguish between grammatical expressions (like language, music, and math) and non-grammatical expressions (like shoulder shrugs and designating makers and categorical tags on blog posts). We could distinguish grammatical expressions in natural languages and formal languages. Though again, "ballad" gets in our way a little. We could distinguish descriptions by what sort of thing they describe, not using domain categories in the full space of nouns, but using a small set of ontological top classes, like situations and activities and quantities. I can do this. It's just going to take some work.
Actually, this might the right time to figure out if I'm going to separately treat types, roles, and instances in the language. Financial objects like "profit", "refund", and "tithe" are roles, I think. I'd like to say that they're roles for quantities, but they feel different from other things I'd call roles for quantities like (limit rating index total tally). Maybe I'm failing to distinguish realized physical quantities from described numerical quantities. Maybe I'll address types versus roles another time.
Okay, I want to talk about things that fully exist at a single time but aren't made of atoms - immaterial countable and massed entities. Let me look through my non-physical categories to see which ones are actually perdurants.
Speech acts like (ban dare oath pact pass plea veto vote vow query fatwa edict ukase order demand denial pledge threat offer promotion prayer) don't exist fully at a given time, but rather across time, so they're perdurants, and in particular, they're countable so they're events. They're distinguished from other kinds of speech in being non-declarative. Declarative speech forms are also events ( gibe joke jest pun quip rant talk tip fib quote recap reply rhyme blurb rumor spiel taunt guess adage maxim brief claim idiom answer remark report saying riddle rumor slogan prompt sermon secret speech eulogy mantra). Massed speech forms are processes/activities (libel gossip praise advice thanks poetry prose), of which the first ones might be communicative and the last two might be narrative? But that probably won't be a distinction represented in the language's phonesthemics. I'm not sure what to call "news". You can talk about things happening during "the news", but that might be short-hand for "the news broadcast", whereas things don't happen during written news. I'll come back to it.
Forms of written composition (novel score haiku fable motto advert resume script survey sonnet poem docket itinerary manifest schedule coupon voucher certificate bill receipt tab manual rubric specification memoir ballot thesis recipe review blog file film list memo menu note pass poll) do fully exist at a time. These things have physical material realizations of course, but we conceive of a novel as being a description (of written content) that physical books can meet. This doesn't mean that the novel exists as a ghost if all its incarnations are destroyed, it just means that the novel's identity is abstracted as a informational pattern from its different physical realizations. I'm not sure how I'm going to treat textual forms that directly specify the medium of realization rather than the content, like (email text letter). Maybe those will be considered concrete (having spatial extent), although I'm not sure whether an email's spatial extent is on a computer's memory or on a display screen. Incidentally, the Xenants are blind and they have tactile display screens.
You can speak of things happening during dances, so countable types of dances are events: (rumba samba tango salsa ballet).
You can't speak of things happening during mathematical objects, so those are in the category of non-material endurants, just like novels. Things like (matrix vector scalar metric ratio prior likelihood proof lemma axiom radius median mean mode integral sum product difference). Is "normal" a type of vector or a role for a vector? Probably a role. I notice roles so rarely, it makes me think that they're either very rare or that they're common and I'm really bad at noticing them. Most processes/activities kind of feel like roles. I'll figure it out eventually.
You can talk about things happening during most musical forms like (opera duet jig raga riff solo tune round waltz carol ditty etude rondo melody shanty minuet anthem medley), so they have senses that are events. But I think that those words have systematic polysemy, with one endurant sense as a timeless specification of how to create/realize a performance and one perdurant sense for a realized performance that matches the specification. So descriptions are endurants, but they can specify the form of endurants or perdurants. Hm. More-narrative artistic forms also have perdurant senses (ode play poem rap saga tale ballad), but I'd say that's systematic polysemy again and maybe the Xenants should have separate words for the description and the described entity, if the English word senses differ in their ontological class, and I want Xenant words to have phonesthemically recognizable ontological classes.
I'd been thinking that units of measure might be a word class for the Xenants and that they might be a subclass of immaterial endurants, but on closer inspection, that doesn't work at all. You can use unit of time words for relative temporal anaphora ("it happened during the (decade epoch moment)") but you can't do the same for units of volume, mass, frequency, capacitance, et cetera. Also, you can use some units of measure as if they had spatial extent, like "I lifted a kilogram", but not so for others, like an octave or an earful. I could say that "decade" has a description-sense and a realization-sense, and likewise with the kilogram. This systematic polysemy is starting to bear a lot of weight, and I'm not sure that it's sturdy enough to do the job. I should refresh myself on what the formal ontographers have to say about it. I'll go read what the Dolce folks have to say about Descriptions And Situations.
Also, I haven't made enough progress on the language's lexicon lately, even if this conceptual work is all going toward constraining the lexicon. Tomorrow I'll try to pick enough noun classes and corresponding phonesthemic prefixes that I can say (the nominal part of) a declaration of rights and the Babel story.
Goodnight.
-
Good morning! While working on a translation of the Babel Text for the ants, I came back to that business of often needing to stick "positive" in front of nouns like (belief feeling declaration desire) when I relate them pertinively to their subjects. I'd be okay with introducing an indicative instrument affix, "Joelle had a belief indicating (john having a dog). The indicative works well enough for (belief, feeling, and declaration), but it makes lese sense to say "Joelle had a desire indicating (John having a dog)", because desires aren't indicative. So how do we positively relate non-indicative cognitive contents to their referents? Instead of using "positive-desire" and the pertinitive affix, you could use a multi-clause counterfactual construction, like "If John were to exist with dog, then Joelle would exist with some appreciation". But I'd like to be able to stick the referent in the verb, the way that "exists with a belief indicating (john having a dog)" would all be one verb. Oh! What about the benefactive affix? "Joelle had a desire for (john having a dog)". Maybe! I was thinking about allowing perdurants to go in the benefactive slot anyway, like inventing a dance for the event of a-gala or causing (water and lime to begin existing as mortar) for the activity of some-building (not that the Xenants are familiar with liquid water). "For use by X" and "For the benefit of X" will both be possible readings of the affix.
Okay, first draft of the Babel Text:
When ants existed as a new thing, the ants existed with a language and the language existed as a unified thing. After the ants existed by means of some wandering, the ants began to exist under a plain and the plain began to exist for the ants as a habitat. The ants caused a declaration to exist indicating that "the ants will cause clay beginning to exist as blocks by means of pressure" and "the ants will cause (blocks beginning to exist as bricks) by means of heat". Then the ants caused bricks existing and the ants caused mortar existing for some building. The ants caused a declaration to exist indicating that "the ants will cause a city and a tower to begin to exist for the ants and the tower will exist with a top at the planet core".
An agent existed at the planet core as a powerful thing and a destructive thing and as no ant. The agent began to increasingly exist toward the plain and away from the core when the ants were causing the city and the tower to increasingly exist as a large thing and a complete thing. The agent existed with perception of the ants existing with the language and of the language existing as a unified thing.
The agent began to exist with a belief indicating that "if the ants exist with a desire pertaining to any goal, then the goal will increasingly exist as a large thing and a complete thing, because the ants exist as a powerful thing by means of the language".
The agent existed with a desire for the ants to not exist as a powerful thing, so the agent caused the language to exist as a divided thing by means of pheromones. Then the agent caused the ants to exist in the planet as as a divided thing in the planet by means of vortices and currents.
While the speaker causes the story to exist, the tower exists an unfinished thing, but the language exists as a unified thing again. The ants exist with memory pertaining to the division. When the ants exist again as a powerful thing, the ants will cause the agent to not exist finally.
The end. Okay, that gives me 37 nouns to categorize ontologically and make words for today. Exciting. I've written the story with periods, but maybe the ants would just keep counting higher with the indexical numerals and call the story one sentence. If not, I'll need to find some way to mark when the counting is starting over again.
Let's have nominal classes come from the (mixed height, mixed articulant) syllables, of which there are 30 (Oi, Ot, Oz, Ox, Ok, Io, It, Iz, Ix, Ik, To, Ti, Tz, Tx, Tk, Zo, Zi, Zt, Zx, Zk, Xo, Xi, Xt, Xz, Xk, Ko, Ki, Kt, Kz, Kx). If (randomly) "Kx" is the material substance class, then "Kx" by itself will be mean "some material" and "Kx" followed by other syllables will be types of materials, or possibly roles for materials or instances of materials. The second syllable in a noun, when there is one, will have shared height and mixed articulant, and again there are 30 such syllables (OI, OT, OZ, OX, OK, IT, IZ, IX, IK, TZ, TX, TK, ZX, ZK, XK, oi, ot, oz, ox, ok, it, iz, ix, ik, tz, tx, tk, zx, zk, xk). Maybe further syllables with == height != articulant will distinguish those 30 materials into subtypes. Nah, that's annoying; things don't naturally divide into 30 subtypes. We'll just have three-syllable nouns be unrelated to their two-syllable prefixes. The two obvious substances in the story of clay and mortar. On earth, clays are defined pedologically by particle size < 0.002 mm or mineralogically by their elemental composition and crystal structure (hydrated aluminium phyllosilicates, usually in the monoclinic crystal system, with some triclinic polymorphs). I think I'll have Kx as a noun class refer to substances including clay minerals and there will be another noun class prefix for composition-agnostic material forms and phases like (liquid fluff powder silt pedological-clay). Clay minerals are formed by weathering feldspar with water, and the Xenants have hardly any water on/in their planet, so clay shouldn't be a short two syllable word for them. But three syllables sound about right. Let's randomly call clay KxOTZK. That's my first word that isn't a number. Cool.
For refreshment, here are the Xenant language sounds, to the best of my IPAing:
O o (pi pu): the bilabial clicks ⟨ʘ⟩ .
I i (ci cu): the dental clicks ⟨ǀ⟩.
T t (ti tu): the voiceless alveolar plosives ⟨t⟩.
Z z (zi zu): the voiceless alveolar affricates ⟨t͡s⟩.
X x (xi xu): the voiceless postalveolar affricates ⟨tʃ⟩.
K k (ki ku): the voiceless velar plosives ⟨k⟩.
So mineralogical clay, KxOTZK, could be romanized long as "kixupitiziki", which might be realized as "kipizi" in the dominant mouth and "xutiki" in the non-dominant mouth. Kixupitiziki looks kind of long and unwieldy for a three syllable word, doesn't it?
I tried recording the word for clay. It doesn't sound good coming from my mouths. What if I offset the mouths by half a beat, so that you got a galloping left-right alternation of syllables? That would hugely increase the distinguishability of sounds, still allow for faster speech than using one mouth, and make it so that one-mouthed humans could kind of speak the language. That feels less cool though, and I want a new language with new things in it, damn it. I just won't record the Xenant words. Or I'll figure out the alien anatomy of articulation and then emulate by tapping a spoon across the mouth of a glass bottle or something and record that into a sequencer, and render words from the sequencer, and just forget about using my meat mouth.
What about the word "mortar"? Mortar is a mixture, but I don't think the Xenants need to distinguish pure substances from mixtures. That'll be in the same category as clay. Again, probably not one of the 30 most important materials, but definitely one the 900 most important materials, so let's make a three syllable noun. I'll just pop over to the browser console for a quick syllables[Math.floor(Math.random() * syllables.length)], and, ah yes, mortar is KxoiZX.
If I were really serious about roles being parallel to types, I could use like Oi for organisms types and Io for organism roles. I won't though. Roles are probably not that important.
Oh gosh, imagine a Xenant trying to say "blood is thicker than water"; "If Xenohemolymph is a viscous thing, then magma is barely a viscous thing".
Okay, back to Babel nouns. There were two processes/activities (i.e. massed perdurants); Some building and some wandering. I don't think the language needs to distinguish processes from activities. Let's have that be a single ontological class. Should the classes relate to each other phonetically? Like endurant prefixes could all have a k or K somewhere and massed prefixes could all have a ... I don't know. I think I have to decide now, because I won't want to go back and change words once I've made them.
My first thought: if the high consonant in the ontological prefix is a plosive, then the word is some kind of endurant, if the high consonant is an affricate, then the the word is some kind of perdurant, and high clicks are reserved for other things. This has potential, but I think I'm going to find it too limiting. So there will be no relationship between Ko and Ok or Ki or whatever. Maybe my next language will have fine grained ontological features represented phonetically. Right now I'm just going to pair things up randomly. That's fine.
Okay, let's address "some building, some wandering". They're processes/activities in English (massed perdurants). The xenants won't make a distinction of massed nouns versus count nouns for perdurants: they'll just have event-like things, and the aspect will be specified on the existential verb. So the difference between "underwent an improvement" and "underwent some improvement" will be "existed in a manner of improvement" versus "was existing in a manner of improvement", I guess. For countable perdurants in the story, we also had "a current" and "a vortex".
If I'm going to have 30 important two-syllable nouns in each ontological class, maybe I should give some thought to what they are instead of just saying that most things are in the less important three-syllable class. What are the 30 most important events/processes/activities to the Xenants?
After trying to pare them down, I'm sure that 30 is too few for the core events, so I'm going to allow the nouns syllables after the first to come from a larger class: Mixed height, mixed articulant are okay (#: 30) and shared height, mixed articulant are okay (#: 30). I'm still working on paring the list down to 60 entries, but I think I can do it eventually.
Okay, still not done, but I'd like to post something here, so I'm going to stop working unlogged. I tried arranging perdurants in antonymic pairs, focusing on the ones that seemed the most simple or ontologically basic or important. I haven't thought yet about which of these perdurants would be better expresses copulatively ("presence" -> "exists-as a present-thing", or which can be expressed by other features of the language like aspect and explicit causation, and I haven't compared the perdurants below to any of the other systems of organization I had previously tried to work out.
All that said, I think maybe the Xenants should have short words for (constancy versus change/variation). And maybe those same words will also cover the meanings of (inactivity activity), (calmness turbulence), (stillness movement), (silence noise), (leisure work), (disuse employment). Or those could be subtypes (constancy change), if I decide to have subtyping speciation. There are also perdurants of the form "cause X to be (all of those pairs)". For example, to cause something to be (inactive active) might be (sedation stimulation).
Maybe the Xenants should have short words for (absence versus presence) or (non-existence versus existence). I kind of like the idea that they equivocate between the two. Non-existent things can exist somewhere in the multiverse or beyond the horizon of the visible universe or whatever; non-existence simply means not currently present. On the other hand, "exist" is one of the Xenants only two verbs, and maybe we don't need a separate noun for it. If you add initiative or terminative aspect to (absence versus presence), you get related perdurants (disappearance appearance), and if you also specify a directive frame then you get perdurants like (departure arrival) and (exit entry). The events (going and coming) are similar, but they have continuative aspect. Maybe they've got the "increasingly" adverbial modifier. "It happened during the coming" means "it happened during the (increasingly-existing-as a present-thing)-situation. Idk. If I do augment the (absence presence) event with a (non-existence/existence) sense, then of course adding a causative gives us (creation versus destruction), and maybe adding aspect onto that gives (initiation termination). If I were doing a poetic thing, I'd have the Xenant - being magmatic organisms - use (melting | solidification) as synonyms for (initiation | termination), but I don't really want them to use metaphors like that.
I've got to go to work right now. I'll write up more later tonight. There's lots more.
-
Well, that was more than a week ago. Some things have gone awry in my life. I don't think I'll be finishing the language this year. But I'll still post occasionally.
I think that perdurants and adjectival nouns in the Xenant language should both come in antonymic pairs. Things that don't have antonyms get put into other classes. And I think it would be very cool if there was a very small tight closed class of root words for both of perdurants and adjectival nouns, and any new perdurants or adjectival nouns would have to be derived from open classes of nouns using derivational affixes.
For example, (movement stillness) can be a pair of perdurants, but manners of motion without antonyms like (climb swim crawl dance quake rotate) will not be considered perdurants by the Xenants. Those will be called plan descriptions, perhaps, and the Xenants won't talk about things happening during them; things happen during motion in the manner of a climb, not during a climb. Having a separate class for manners seems better than making up words like "unswim" and "unquake". Maybe there are also manners of contact like (touch hit squeeze throw cut puncture break breach stab). And some cognitive events like (computation/calculation choice/decision/selection deduction discovery judgement/categorization) don't have clear antonyms, so how about we call those manners of cognition (or roles for computation or something). Some communicative events (statement speech demonstration display indication deception) don't seem to have antonyms. I think it makes sense to call those roles, rather than manners, but I don't have a good understanding of the difference. Maybe (measurement test search/hunt exploration) are roles or manners for active perception. And then a bunch of events where you can say "there was an X between the two of them", like (exchange/trade competition/fight/race/chase conversation/discussion/consultation) might not have antonyms. Although (agreement disagreement) form an antonymic pair that can be had between two people, so what's with that? And reconciliation? I bet there's an antonym for that too. There are some events like (offer, bet, contract) that don't have antonyms, but we can just treat those as "giving an X" or "communicating an X", and I think it makes sense to call all three of those things "plan descriptions" again, so that's working swimmingly. I think (lottery raffle, auction, sweepstakes) might be types of offers, and maybe they can be treated similarly in a non-perdurant way.
I think the antonymic perdurant pairs I do have would be better presented as a list than with exposition, as I had been doing, so here they are:
: (sedation stimulation)
: (failure success), (defeat victory)
* (absence presence)
: (appearance disappearance), (arrival departure), (entry exit), (coming going)
* (creation destruction), (initiation termination)
* (decrease increase), (gain loss)
: (forgetting learning/remembering/perceiving)
: (degradation improvement)
: (intake/absorption emission/expulsion)
: (taking giving)
: (assistance attack), (hostility hospitality), (reward punishment)
* (division unification)
That was me trying to whittle my list down to 60 base perdurants (30 pairs) that would get short words. Maybe there should be a pair for possession/lacking, but you don't usually say "it happened during the lacking". Although you do say, "it happened during the (shortage | famine | scarcity)". So maybe I'll add it. It's more stative than eventive, maybe? Most of the things above are changes. Idk. There were a few other things I had trouble categorizing. A voyage/journey/migration/leave seems kind of antonymic to a visit/stay, but those also feel different from the antonyms above somehow. Maybe those are roles for absence/presence.
The set of (placement, alignment, allotment, allocation, distribution, rearrangement, reordering) feels pretty semantically basic, but they all seem roughly synonymous, and there's no antonym, and there's no category of similar things the way that the manners of motion formed a category of similar things. So what do I do with that? I can represent "ordering" as "causing it/them to have some order", but how do I represent "reordering"? I think I need to treat "orderings" as distinguishable countable nouns. But they're not descriptions of plans/behavior, they're descriptions of state. And I was going to represent states as "exist-as $adjectival-noun", like "exist-as ordered-thing". And those can be countable, so that's fine. But if adjectival nouns are state descriptors, should I have adverbial nouns for plan/behavior descriptors like "crawl" and "swim"? I don't want to think about it right now.
I might want a simple Xenant word to cover the meanings like (binding liberating) or similar things like (freedom slavery), (capture release), (escape imprisonment). Freedom/slavery is stative and the others aren't, but whatever. I wasn't sure about some of those antonyms. "Evasion" might be in there somewhere too. I had worked on representing the event semantics at a finer granularity to figure out the right antonymic pairs, but I can't read my handwritten notes from that day, and I've forgotten what I wrote, so that'll be something to redo on a later date.
I put (assistance attack), (hostility hospitality), (reward punishment) above, but they sure seems more role-like than most of the others, do they not? The same way that just about any physical activity can play the role of an exercise, just about any interpersonal activity can be assistance or interference. I should be more systematic in my investigation of roles. Here are some other ones: (cause, effect, problem, solution, job, leisure, simulation, reaction, distraction, stimulus, attempt, crime, achievement). Lots of perdurants can play the role of those things. If you have a theory of perdurant roles that neatly organizes those, I'd be happy to listen. I know that the Guarino / Welty / Gangemi crowd has written about the ontology of roles, but I read it and it never sticks in my head, which might be because of the limited/unclear presentation, or it might be because it's just not conceptual machinery that I find useful/correct enough to adopt.
How about adjectival nouns? I'd like those to be antonymic too. And I think I can make a pretty small and tidy set. Hopefully I'll post that soon. Probably not tomorrow. Maybe next Monday or Thursday or something. Everything is falling apart. Goodnight.
-
The Xenants won't have adjectival nouns corresponding to English modal adjectives of possibility like (acceptable accessible comparable comprehensible compressible convertible corruptible countable defensible deliverable destructible digestible discernible disposable divisible edible exhaustible expressible extendible fallible flammable legible moveable perceptible perishable portable profitable reliable sensible), which a thing "can be $verb-ed" or "can be caused to $verb". They don't see those as properties of things. If a person can't eat some kind of rock, that's a property of the person (or maybe the agent-patient combination in some cases) and not of the rock. If you want to tell a Xenant that a conference room is available, try telling them that it is empty instead. If I were to have modal adjectives, the ones I'd want most would be (avoidable unavoidable) and (achievable unachievable). But no, I shall try to do without.
Okay, I'm down to a little more than 100 pairs of adjectives that I think the Xenants would have adjectival nouns for. That's not 30 pairs that can fit into 60 two-syllable words, but it's a good start. I'll present them organized into domain categories, but at some point I hope to organize them by a better scheme, like based on what sorts of objects the adjectives can modify.
Cognition: (intelligent unintelligent) (knowledgeable ignorant) (attentive inattentive)
Other biology: (live dead) (conscious unconscious, awake asleep) (immature mature) (sick healthy) (energetic exhausted) (sober intoxicated) (sensitive insensible). For these and the cognitive ones, I'd also be fine with just having regular nouns that would allow me to say that a Xenant EXISTS-with (intelligence, knowledge, attention, life, consciousness, maturity, illness, energy, toxins). They might also have (hungry fed), although I'd rather use a pair that isn't so specific to biology like (unsatisfied satisfied) or (supplied under-supplied). I don't mean emotional satisfaction, I just mean satisfaction in the sense of meeting a standard, like how constraints can be satisfied. The (sensitive insensible) pair is used to talk about functioning of sense organs, instead of treating each sense modality categorically with adjectives like (deaf, blind, numb, anosmic). By sober, I don't mean "abstaining from recreational toxins", I just mean free of toxins/pheremones. There's probably a better word for it. "Clean" as an addict, or "dry" as a county, or ... uncontaminated.
The Xenants have a very limited repertoire of emotions and personality traits, however I think they would have: (confident uncertain) (autonomous managed/enslaved) (creative destructive) (adept unskilled) (reputable disreputable), and some other things that aren't necessarily applied to causal agents like (ready unprepared) (active inactive, constant variable). I'm not sure, but I don't think they have words like (well-behaved misbehaved). They don't expect people to be well behaved, they expect them to be (reputable disreputable). And maybe they don't have (tolerant intolerant) or (strict lax), because they're only strict and intolerant? Idk. They're definitely awful little monsters, bereft of human sympathy.
Ontological adjectives: (abstract concrete) (singular plural). I'm not sure about these. The xenant language doesn't mark for singular/plural, so they might not care about having adjectives for that. And their language *does* mark for abstract or concrete in the ontological phonesthemic noun classes, so having adjectives for those feels redundant? Those arguments aren't compatible: I can't skip a thing because it's not marked and skip it because it is marked. Let's keep both pairs for now.
I also thought about including (endurant perdurant), but those are nouns rather than adjectives, and ... that's fine. I also thought about having antonymic adjectives (countable massed). Maybe to make something be massed is to grind it and to make it be countable is to lump it? Sure, they have adjectives for (countable massed).
Perceptual adjectives: (perceptible imperceptible) is the modality-agnostic pair. Oh wait, no that's one of those ones like "flammable" that's agent specific. No modality-agnostic pairs, I guess. You just ask whether the agent is sensitive or insensible with respect to the thing. For modality-specific adjectives, the Xenants can definitely hear, so I want (faint loud), and they can feel, so let's have texture pairs of (rough smooth, coarse fine) and (hard soft, rigid flexible). I like the idea that they can sense particle radiation, especially if that's the energetic currency of their ecosystem, so maybe they have dedicated adjectives for (radioactive radio-inactive). For thermoception, they have (hot cold). I think they also have words for (reflective non-reflective) or (clear opaque), but I'm not sure if they're inferring those as electromagnetic properties despite being blind, or if the adjectives are being used for particle radiation (like opacity means radiation-stopping power). The Xenants also have modality-specific affixes for deriving sensory adjectival nouns from nouns in the open classes, like a thing feeling (stony, metallic, creamy, ...), or tasting of (sugar salt garlic tannin acid ...). Oh hey, I don't have any root adjectives for taste. How about things taste (caloric non-caloric)? Mmm, not sure.
For spatial adjectives, we have size adjectives (large small). Dimensions include (thick thin, narrow wide) and (long short) and (tall/deep shallow). Why does English use "short" as an antonym to both "long" and "tall"? I honestly thought about combining long with tall/deep and having "short" be the antonym for both just because English has fucked me up that badly. But no, we live in three dimensions and so they shall have adjectival nouns for three dimensions. For shapes, let's have (solid hollow) (bent straight) (dull sharp). I kind of want to have one word for (level uneven) and (bent straight), but that might conflating shape with orientation? I'm not sure how to handle orientations at all. The pair of (upside-down right-side-up) seems basic, so it's weird that they're phrasal in English. They'll each be one word for the Xenants. The pair (horizontal vertical) make a nice pair in English, but the Xenants live inside a planet, so they don't have a horizon and "horizontal" is probably not the right word for the concept they'd come up with. How about they orient things (perpendicular parallel) to local lines running through the planet core. I should probably base the locative adjectival nouns on the locative verbal frames, but on my first pass of just looking at adjectives and not thinking about language features, I just wanted (present absent), (internal external), (distant near), (central peripheral). I also thought that (distant near) could be a synonym for (opposite adjacent), but maybe that needs revision also. I'm not sure the Xenants care about places of origin, but if they do, then they can have (indigenous foreign). For configurational adjectives that relate the position of multiple things, let's have (open closed), (full empty), (organized disorganized), (covered bare), (joined separate, united divided). I think (covered bare)-thing would get used in a verbal construction with an open class noun like "filth" to cover for the meaning of the pair (clean dirty), if the Xenants even have a concept of filth. They do have a concept of impurity, and maybe (full empty) of impurities is how they represent the dimension of (pure impure). Maybe (perfect flawed) will work in the same way. Or, no, flaws aren't necessarily spatial, so we shouldn't place them with the spatial adjectivals (full-thing empty-thing). But (sober intoxicated) can work that way. What about (damaged undamaged)? I think they'll conceive of damage as existing in space, but "full of damage" is still not how they'll express that. I'll have to come back to that one.
For temporal adjectives, they have (new old), (early late), (brief lengthy), (frequent infrequent), and maybe (continual intermittent) (regular irregular) (permanent temporary). The (frequent infrequent) pair is used for perdurants (e.g. events, states, processes), whereas you would use (normal unusual, common rare) to talk about typicality of endurants (like objects, substances, and locations). The "old" in "new old" does not carry any implication about biological function or dysfunction, as it does when English speakers contrast "old" with "young". The Xenants have (immature mature) for that.
I don't know if I want aspectual, modal, or epistemic adjectives roots in the language. Maybe (ongoing non-ongoing) (beginning ending) (complete incomplete, successful unsuccessful) for aspects. To match the verbal modalities, they should just have (hypothetical real), and not like (possible impossible) or (necessary unnecessary). "Possible" is already one of those adjectives I ruled out at the beginning of today's writing like "flammable". Different things are possible for different agents. I do kind of still want (necessary unnecessary) as a pair though, but maybe I'll put it in a non-modal domain, even though logicians like to represent it with modal logics. For epistemic adjectives I was thinking of having (probable improbable) (known unknown/secret, found lost) (obvious/apparent subtle). I wasn't thinking about Xenant evidentiality when I decided on those though. That could use some revision. Hm, the pair of (obvious/apparent subtle) feels like a near synonym for (perceptible imperceptible), which I decided against including in the adjectival roots. If there's a sense in which obviousness is intrinsic to an object and not to an observer, that sense is probably a synonym for simple/complex, no? Maybe not.
Adjectives of force dynamics: I'd like the Xenants to have adjectives for directed force: (tense/strained relaxed) and (tight/compressed loose). The pair of (compressed loose) feels really close to (bound free) which I don't have categorized anywhere, but (bound free) also feels close to (joined separate, united divided). which were in the spatial configuration adjective section. So that's an interesting bridge of near-synonymies. For other forces, I don't like the English pair (strong weak). The Xenants will split that into (powerful feeble) and (tough fragile). They have adjectives related to hefting (heavy light), and they use these kind freely to talk about mass, and weight, and density, just as we do. Maybe those should have gone in the perceptual adjectives section, but this is fine too. Multiple inheritance is a hazard of working with domain categories. For motion, they have adjectives of (motile sessile, mobile immobile) and (slow fast). They also have words for (stable unstable). I'm not sure where to put (mild extreme/intense), but this seems like an okay place. They have words for (mild extreme/intense). I think the Xenants might use (powerful feeble) to express (effective ineffective). Those are pretty close.
Mereological adjectives: (whole partial). That's it. Although I'm not sure I even need that, if I have (joined separate, united divided). Let's keep it for now. And maybe (complete incomplete, successful unsuccessful) from the aspectual adjective domain category should go here instead. Endurants are whole, perdurants are complete.
Properties of symbolic objects, messages, measurements, thoughts: (specific general, precise vague) (concise verbose) and (simple complex). Where English speakers use "long" for a spatial dimension, temporal duration, and abstract symbol counts, the Xenants use three words ("long", "lengthy", "verbose").
Adjectives of correspondence: The Xenants have lots of logical adjectives (true false) (valid invalid) (sound unsound) (consistent inconsistent). They also have (equivalent inequivalent, equal distinct). Maybe (similar dissimilar) are expressed by putting a diminutive affix on (equivalent inequivalent). Equivalent in some respects. Other compliance adjectives: (compliant non-compliant) (uniform diverse) (dependent independent) (direct indirect/meandering). Maybe (relevant irrelevant, related unrelated) also goes here. And (necessary unnecessary) could go here too, I guess. This category is admittedly kind of semantically heterogenous. If I have the (dependent independent) pair, maybe I don't need (autonomous managed/enslaved) in the personality section. Maybe (compliant non-compliant) covers some of the meaning that I was trying to convey with (unsatisfied satisfied) or (supplied under-supplied)? Maybe not enough.
Evaluative adjectives: The Xenants have (good bad). (important unimportant), (sufficient insufficient, adequate inadequate), (easy difficult), (safe risky/dangerous). I think they have (helpful harmful) but they don't use it as a personality descriptor, because Xenants are monsters, bereft of human sympathy. The Xenants also have words for (valuable worthless, valued despised). I'm not sure if they have separate words for (expensive inexpensive). Probably they do, but there's an opportunity for world building / psychological characterization in having the concepts conflated. There are a few more evaluative adjectives, that might be reasons for a thing being valuable? These are that the thing is (profitable unprofitable) (efficient inefficient) (attractive ugly) (functional non-functional, capable incapable) (useful useless, practical impractical). Hm, I really have no basis for separating most of these from the behavioral trait adjectives at the start. Oh well. Oops, I think the pair (profitable unprofitable) is in the same category as modal category as "flammable" that I didn't want to have roots for. Let's get rid of that pair.
The Xenants don't distinguish between (natural artificial) or (original duplicate). If they have a sense of things being (public private), then it's very, very limited compared to that of humans. I don't think they care about (fairness injustice). I think they use some other word class besides adjectival roots to represent concepts like (initial final), (latter former), (previous next), (past recent present current future), (delayed immediate sudden), (primary secondary tertiary ...), (annual weekly daily hourly minutely ...). I'm not sure they have adjectives for parts, like (upper lower)-thing. They would just say that X exists at the top of Y, for example. We'll see what happens after I rework the locative adjectives to reflect the locative verbal affix system.
The Xenants have quite a few derivational affixes that apply to open-class nouns to make adjectivals like:
* Pertaining to physical domains like (electricity magnetism chemistry ....) and other domains like (industry monopolies democracy spirituality cognition perception ...), giving adjectival nouns like electrical-thing and industrial-thing.
* Located in or operating in a (city country suburb nation ...),
* Possessing (clothing money courage beauty ...) or being covered in (mountains scales ...) or containing X or being made of X or being free of X.
* Being in the style of various creative movements or people.
And others.
Okay, those are the antonymic adjectival roots. Is it worth my time to try paring them down into a set of 30 pairs that can fit in the space of two syllable words in an ontological category? Probably not. They're already pared down as much as I want, and there's about 112. If I could split them neatly in quarters, giving four groups adjectives, each with 30 pairs. Maybe there could be endurant adjectives and perdurant adjectives. Except... only like 10 of the pairs are actually pertinent to perdurants. Maybe if I try nominalizing the adjectives, I'll see some patterns there.
If I were a more competent ontographer, the events would either have a parallel structure to the adjectives or they would have no semantic overlap, but they wouldn't have semantic overlap and differing structures as they do now. That needs some work.
See you later, folks. ~ Love, James.
-
I tried nominalizing those adjective pairs. I noticed that the pair of (internal external) rarely gets nominalized in English: we speaker of internality and externality economically, but almost never locatively. It almost makes me wonder if the pair of (internal external) should be removed from the set of adjectival roots, but I don't really see how it differs semantically from (present absent), (distant near), or (central peripheral), which have nominalizations. Just a quirk of English, I guess. Horizontality and verticality are also pretty rare in my speech, but they're fine too, I think. Perpendicularity and parallelism are fine. Are there nominalizations for (upside-down right-side-up)? We caused it to have some up-side-down-ness? Maybe (orientation disorientation) work as nominalizations there? Kind of. Oh, we have "uprightness", although it's usually used in contract to crookedness rather than upsidedownity / overturnedness / turtle-turnedness / keelhauledness / capizicity / capsizement / pitchpoledness / upsetedness / .... inversion! Oh my god, I was seriously concerned there was no common English word for it. Uprightness and inversion. Good enough. Something weird is still going on here with the location and orientation words mostly not having good/common nominalizations.
I think I might have made a mistake including (helpful harmful). They feel kind of antonymic, but I think I'd rather split up the pair into semantically more granular (helpful unhelpful) and (harmful non-harmful). The first pair are synonymous with (useful useless, practical impractical), which we already have, so let's get rid of (helpful unhelpful). Oh wait, is useful just usable in disguise? Maybe it's one of those adjectives like "flammable" that I don't want. Not sure. I think the second pair, (harmful non-harmful) are synonymous with (safe risky/dangerous). I like the (harmful non-harmful) labels better, so let's remove (safe risky/dangerous). I probably made the same mistake when talking about perdurants ((assistance attack), (hostility hospitality), (reward punishment)).
Maybe the Xenants don't have general words for (good bad). They instead use more fine grained dimensions like things being (useful, valuable, attractive, important, pure, powerful, ...). I like that.
I kind of want to remove (necessary unnecessary) from the adjectival roots. Maybe those can be expressed with an augmentative affix operating on (useful useless). But (necessary unnecessary) forms some kind of pair of pairs with (sufficient insufficient), and I don't know how to express (sufficient insufficient) another way. It's not expressible as with a diminutive affix operating on (useful useless) for example. There's some scalar thing going on the pair of pairs, right? Necessary means "need at least this much" (like >= 3 dogs) and sufficient means "need no more than this much" (like <= 3 dogs). I guess keep both pairs of pairs for now. The Xenant's ability to express comparatives is poor enough already without removing those.
I didn't mention the pair (broken intact) before, and I think the Xenants don't have it: they talk separately about things being (whole partial) or (united divided) from things being (functional non-functional).
I'm removing (known unknown/secret, found lost) pair from the adjectival roots. Those are too much like the "flammable" adjectives that Xenants don't have. Things aren't characterized as being known, but rather an agent is characterized as knowing about things. Let's also get rid of (obvious/apparent subtle). Xenants will instead talk about reasons why something might be obvious or subtle, like the intensity of a stimulus or the abstractness of a concept or the commonness of a belief or whatever.
Almost all of the adjective pairs can apply to physical objects like a dog. There are a few that I want to be specific to non-physical things like statements and measurement. Here they are, presented with one adjective from each pair nominalized as a dimension: (validity truth specificity soundness probability concision). The dimensions for perdurants will include (brevity frequency regularity intensity earliness continuity direction). I'm still not sure if I want the aspectual (complete incomplete, successful unsuccessful) or (ongoing non-ongoing) as pairs of adjectival roots for perdurants, but that's probably due to my dissatisfaction with the labels. There should definitely be some way to talk about the (existing, happening, transpiring, appearing, prevailing)-ness of perdurants at a time. This has been a little bit of a peeve of mine since I was a kid: whenever a commercial said that a sale was "going on now", I got frustrated that there wasn't a better common way to express existence of perdurants than with a phrasal verb that included a locative adposition. You should just be able to say, "our sale prevails!" and people would get excited if they believed you, which they won't, because you've been broadcasting the same t.v. advertisement continually for years, you silly inarticulate conmen. Um... yeah, fine, perdurants can have (completion and currency) alongside (brevity frequency regularity intensity earliness continuity and direction). A further thing I don't like about "current" is that it's kind of cheating in its binarity; (past, present, future) are a categorical distinction, not binary one, and having (present versus (past and future)) is like sorting all nouns into apples and non-apples. It just doesn't cut at the joints of reality. But it does cut at the joint of experience, I guess, and there's nothing wrong with language having a cognitive bias. So yeah, "current" for the Xenants isn't an antonym of past, it's an antonym of non-current. Or, idk, maybe I should just have "incomplete" imply current.
My list of adjectives for physical things like dogs has just 72 pairs at the moment. Here are the dimensional forms (nominalizing one word from each pair): (activity age attentiveness autonomy beauty capability centrality coarseness complexity compliance confidence consciousness coverage creativity dependence distance efficiency employment emptiness energy equivalence existence fragility hardness harmfulness health heat heaviness importance intelligence knowledge largeness length life loudness maturity mobility necessity normality opacity openness organization orientation permanence plurality power preparation presence radioactivity rarity reflectivity relevance reputability rigidity satisfaction sensitivity sharpness skill smoothness solidity speed stability straightness sufficiency tension tightness totality uniformity unity usefulness value width). If I can find a property that splits that list in roughly in half, I think I can get rid of 12 pairs and get two small ontological classes of 30 pairs each. I might not be able to do that, but maybe. It would be nice.
Wait, that list doesn't have perpendicularity/parallelism. I messed something up.
Well, let's plow ahead anyway and clean up later. A nice thing about having just one word of each pair nominalized, is that I can move one word around more easily in a text editor than a pair of words, which lets me come up with new categories/clusters quickly.
Here are some new ones:
Spatial stative: (presence unity openness hollowness emptiness coverage neatness centrality distantness size length width height straightness invertedness sharpness smoothness fineness tension tightness heaviness hardness rigidity opacity hotness)
Abstract stative: (complexity hypotheticality dependence sufficiency reputability relevance value necessity equivalence rarity importance normality intensity compliance preparedness uniformity wholeness plurality newness attractiveness knowledgeability confidence maturity)
Dynamic: (powerfulness capability creativity efficiency stability fragility skilfullness sensitivity activity radioactivity mobility permanence autonomy speed reflectivity harmfulness intelligence attentiveness consciousness healthiness liveliness)
I'm not sure where "employment" should go, as contrasted with disusedness (yes, I know it's not a word). If I try to come up with a synonym for "employment" in hopes that it will be easier to categorize, the word my mind first tries to construct is "participatoriness", which again is not a real word. English has the word "participation" of course, but I'm not sure that's good enough: it's starting to seem to me as I look at these words too much that "attractiveness" means something a little different from "beauty" and "healthiness" means something a little different from "health" and "hotness" means something a little different from "warmth" or "heat". Maybe employment is one of those adjectives like "flammable" that I wanted to skip? Like .. whether someone considers your existence to be useful doesn't necessarily say anything about what you're doing. And if you're actually doing something, we already have the adjective "activity", so ... does having "employment" give us anything? I shouldn't just categorize words as being in the ignorable category when I have trouble categorizing them, but I think I've made a good case for not using this particular difficult word.
I'm also having trouble categorizing (satisfied/supplied), maybe because those two aren't that closely related but I'm trying to think of them as one thing. Let's see. Supplied is like being equipped. Is there a nominal form of that? No, not really. It's also like being endowed, for which we have the noun "endowment", although maybe that nominalized the verb, rather than the adjective based on the result of the verb. Um..."richness" and "affluence" kind of work, except I just want to talk about possession, and not necessarily valuable possession. I'd probably be happy with "satisfaction" as a nominalization if we didn't use that so heavily to refer to an emotional state in English. "Satisfaction" feels more abstract-stative to me and endowment feels more spatial-stative, probably because I'm thinking of endowment as physical possession.
Maybe these new categories need a little tweaking of the members for coherence, or they need definitions to demonstrate that the memberships are valid. And maybe the "dynamic" category should be split into abstract and spatial varieties for the sake of symmetry, or something. I'm not sure. But I think I'm getting nearer to finding some useful and regular conceptual distinctions that separate my core adjectives into moderately large clusters (of 20 to 30 pairs each), and that's nice. That's a good level of semantic granularity I think and also it plays well with my phonology, if I ever get back to making up fake unpronounceable words.
Next I'm going to take a short break from adjectives to try cross-referencing the perdurant pairs I have above in this blogpost against an early perdurant organization scheme of mine that I never shared. Once I've got that cleaned up, maybe I'll see some relationship between perdurant organization and adjectival organization. Goodnight.
-
I like the adjective categories from yesterday a lot, but I'm struggling to find examples where I can use adjectives from one set but not another, and I don't want to draw a distinction between the adjectives if the distinction doesn't influence language usage at all. If (hollowness, sufficiency, stability) come from three classes of adjectives, then I want three classes of nouns that they can modify, or something like that. Categorization should have consequences. Philosophical distinctions worth using should help you to regularize your grammar.
Ooh! I've had three points of conceptual progress suddenly.
First, I didn't want the language to have privative adjectives/adjectivals like (fake, phony, toy, counterfeit, past) - the sort where in English (fake dog) implies it's not a dog. I'm not sure if (imaginary hypothetical) are on that list of privative adjectives. Is a hypothetical rock a rock? We make a tremendous number of inferences about a hypothetical rock like it is a rock. I'm not sure. If "hypothetical" is privative, then I'd want to get rid of the pair of (real hypothetical). Xenants would just have to use verbal mood to talk about hypothetical existence.
Second point of progress: all of the adjectives in the three categories from yesterday can be applied to concrete things like dogs, but only some of them can be applied to non-material causal agents like computer programs / artificial intelligences. So if we conceive of (broken intact) as applying to material objects, then the pair should be in the first (spatial) class of adjective and not in the third dynamic class. Or better yet, split the dynamic class in material-dynamic and immaterial-dynamic classes, and and put (broken intact) in the material one. If you have an immaterial concept of breaking, that one can go in the immaterial dynamic class. Maybe a fragile AI is one that becomes non-functional if you make small changes to its code or if it makes a small departure from its normal state space. Likewise, "speed" conceived of as a rate of change of position should be in a material category, whereas speed of a computer program would be in the dynamic immaterial adjective group. A quick tornado will have a different adjective than a quick computer program. Although maybe quickness for a computer program means time-efficiency or IO-efficiency or just plain efficiency.
I have a few worries about the philosophical grounding of introducing the material-immaterial distinction on adjectives; First, the state space of a Turing machine is a kind of space, if not a continuous metric space like our three spatial dimensions, and when I call a program "fast in an immaterial way", it feels close to calling it fast in a non-spatial way, and I'm not sure that's accurate. A speeds involve time and some kind of space. So I'll start saying (material immaterial) instead of (spatial abstract).
Second, when you talk about a computer program being responsive, you don't necessarily know whether the code is efficient or whether the computer is powerful. But a powerful computer is fast in a material way, because matter is moving through metric space. So it seems that there are common situations where a speaker, wanting to comment on a speed, won't know whether to use a material adjective or an immaterial adjective. Maybe a Xenant would say that the user interface is materially fast, and then ask whether the processor is materially fast or the program is immaterially efficient? That seems like it works, but like, what's really gained by separating "speed" into material and immaterial concepts? It seems more like I'm regularizing my word lists than regularizing the grammar. But I like regularity and semantic uniformity and symmetry and stuff, so let's keep the distinction for now.
Third point of progress: Maybe the adjectives I was calling stative can be used to describe things that never change, whereas the dynamic ones can not. You can be a powerful thing and not be changing at the moment, but you can't be a powerful thing and never change. I'm not sure how useful this conception is for categorization in practice. If something is ready/prepared, it could stay that way forever after, which seems stative, but maybe the readying / preparation, as a change, means that the adjective should be dynamic. Unless something can be ready for use without having been readied. Like, we can conceive of a complex thing existing unchanged for all times without ever having been complexified / complicated, so "complex" is a static adjective, even though "complexification" is a perdurant we can conceive of. So following the example of "complex", maybe the Xenant concept of "ready" doesn't mean "having been readied", but simply "not needing to be readied". That might be a synonym of "usable" in some cases, like an oven being ready for use, and in other cases, like an agent being in a state of readiness, it might be a synonym for "eager" or "equipped", or some more general adjective encompassing both like "possessing of necessary resources, mental or material". (Ready unprepared) are kind of weird pair, because they require a context: you're ready for something or ready with respect to something. But I think I'm okay with that: if you can be skillful with respect to surgery, then you can be ready with respect to surgery. I'm less sure that the non-agentive use of "ready" is suitable for the Xenants: "usable" sounds like one of those adjectives like "legible" and "edible" that are more about the abilities of an agent than about the properties of the patient. Maybe instead of saying "prepare ye the way of the lord", the Xenants would just say, "make straight a highway for god"; they would talk directly about what situation they wanted prepared.
Okay, I'm going to reintroduce the adjective pairs instead of working with one nominalized word from each pair, and I'll see how the (static dynamic) and (material immaterial) distinctions influence the domain categories that I had for those.
I think I'm not going to have the Xenants conflate existence with presence, even though that sounds cute and poetic and alien. Presence is a material static adjective while existence is static, but suitable for material and immaterial objects, like dogs and sonnets. I will not conflate words in different ontological categories so willy-nilly. Oh shoot, what about the adjective pairs (material immaterial) and (static dynamic)? I can't fit both words of either pair into any category. Hm... although (material immaterial) are both static. Hm... Oh, but when I say "material adjective", I'm categorizing adjectives based on their ability to modify material nouns, not based on the materality of the adjective. The antonymic adjectives that apply to adjectives could better be called (material-concerning not-necessarily-material-concerning), both of which go in the immaterial category. There might still be a problem with categorizing (material immaterial) as they characterize non-adjectives though. Hm... if I gave on categorizing an adjective and its antonym in the same category, then ... Hm... If you can't have a material AI or an immaterial dog, then ... Hm ...
There are a tiny few adjectives in the current Xenant inventory that I think of as being non-gradable or non-comparable. As I usually conceive of the concepts, it doesn't make sense to ask which of two things is more real or more singular, for example. Also there are some few adjectives where an entity can only possesses the corresponding trait with respect to another entity, like you can't have something be equal by itself, but only equal to another thing. Likewise for relevant, dependent, and compliant. Non-gradeability and relative-existence are interesting features of adjectives and I'm tempted to subcategorize all the adjectives based on those properties, but they're so rare, I'm not sure it's worthwhile. Also the evaluative adjectives have something going on where X is valuable (for / with respect to) a valuator, which I think is slightly different than the (equality relevance dependence compliance) thing. When you "cause X to have some equality with Y", you're changing X, but when you "cause X to have importance with Y", you're changing Y? No, not necessarily. I'm not sure how to articulate the difference, but I do think there is a difference.
...
-
It's been a while. I've started working on an ontological categorization of nouns for the Xenants. I should finish the adjectives first, but I didn't and I won't.
I'd planned to have nouns be very very free and open and inclusive in the Xenant language and have other parts of speech very very restricted. Now I'm wondering if I should be more disciplined or ascetic in my inclusion of some Xenants nouns. For example, if I'm getting rid of verbs except for EXIST and CAUSE, it's kind of cheating to have nouns like (amplifier, mixer, wiper, sawyer), is it not? Since they're derived from verbs? Maybe not. I think I'd like to try not having such nouns at first. Maybe I'll let them back in later. Far and away the two largest ontological classes of nouns I have so far by membership are Artefacts and Agents, and lots of those are verb-ers, for example agents in five letters include (biter coder diver fixer giver hater idler joker lover maker ogler payer racer skier taker urger voter wader). Those will be much smaller classes after getting rid of $verb-er nouns. I might put the non-endurant and non-adjectival nominal ontology in a further post and keep on working on adjectivals and endurants in this one (when I get back to them).
I think I'm okay with ... having an affix that derives a word like "saywer" as "user of a saw", and I just want to avoid derivations like "one who saws". And even derivations from endurants are okay, like "one who causes division", if division is to be an endurant root of the language. Basically, I'm going to start with perdurant roots that look more like "saw" and "vote", and then maybe later I'll figure out affixes that produce from those some new words like "sawyer" and "voter". Maybe "saw(-user)" or "vote(-producer)", but not "sawing-doer" or "voting-doer". I guess. It's tricky because "sawing" and "voting" are also activities by zero derivation, and activities are nouns. Maybe I'll try to avoid deriving perdurants from endurants too. See how that works. Probably it won't work well. I can deal with that.
No comments:
Post a Comment