Agents, Patients, and more

I was thinking this morning about the noun view in English, and what it means in a sentence such as:

My hotel room has a great view.

In this sentence, view is describing what is seen, or what is viewed. English is also capable of using nouns derived from verbs to indicate the agent, most often by the use of -er, or even the object of the action, sometimes with -ee, though this is a bit rare. For example, a payee receives funds from the payer. This -ee suffix is used fairly productively in colloquial speech, at least in certain parts of the US, even when the resulting word is never used in print or on official documentation. Theoretically, at least, one could take any verb and attach -ee to make a noun denoting the object of the action. In such a case, the verb that we take would be the takee.

Considering this made me think about conlanging, and how it might be used to explore such an issue, which is precisely what this blog is supposed to be about. Let's begin with a further explanation of the English case, particularly the agentive -er suffix.

This suffix is in fact, a little ambiguous at times, because viewer could technically mean the person or other being doing the viewing, but it usually refers to an instrument of some kind used to view specific things, like a kaleidoscope viewer, or software for viewing pictures, etc. Why the ambiguity? The short answer is to shrug our shoulders and chock it up to the fact that, sorry, it's just English, which is a very typical explanation I give to my ESOL students.

This is where conlangs could come in. Let's imagine a conlang that derives nouns from verbs corresponding to several semantic roles. This language will have a marker for agent, patient, theme, and instrument, which correspond roughly to the syntactic roles of subject, direct object, indirect object, and well, instrument, respectively. Each of these will be indicated by a different suffix (I suppose if we really wanted to be cool, we could use prefixes, infixes, or some sort of root mutation instead, but we'll keep it simple for now).

Let's use the following suffixes:

agentive: -(e)r

patient: -i

theme: -(e)n

instrument: -u

For the fun of it, and to make look a little less like English, let's make verbs by spelling them backwards from English, with simplified sounds and letters. In other words, we'll reverse the spelling of English verbs and turn complicated sounds like th, sh, ch, q, and x into simply t, s, s, k, and k, respectively. We'll also ignore any silent letters like k in kn, w in wr, and the silent e on the end of so many verbs. So let's take about 10 common English verbs and apply these transformations:

1. be - ib

2. see - is

3. go - og

4. write - tair

5. think - nit

6. do - ud

7. sleep - pils

8. love - vol

9. want - naw

10. have - vah

So, attaching the suffixes we would have the following words:

1. iber be-er, being, person

ibi what is, being, existence

iben something to or for which things exist

ibu something by or with which things exist

2. iser seer, someone who sees

isi something that is seen, a view, or perhaps a picture

isen something to or for which one sees

isu something by or with which one sees, viewer, spectacles, a telescope, etc.

3. oger someone or something that goes, goer, perhaps car or other vehicle

(ogi) not used, as go is an intransitive verb

ogen a person, thing, or place to or for which something goes

ogu something by or with which one goes, a vehicle, car, etc.

4. tairer someone who writes, writer, author

tairi something that is written, literature, a poem, or written story

tairen someone or something to or for which one writes, addressee of a letter, theme or purpose of a work

tairu something by or with which one writes, pencil, pen, or other writing instrument

5. niter someone who thinks, thinker, intellectual, scholar, philosopher

niti something that is thought, a thought, idea, belief, or philosophy

niten something or someone to or for which one thinks, the cause or theme of thought, the object of thought (perhaps especially a loved one)

nitu something by or with which one thinks, the mind

6. uder someone who does or acts, actor, agent

udi something that is done, action, act

uden something to or for which one acts, goal, purpose

udu something by or with which one acts, will

7. pilser someone who sleeps, sleeper

(pilsi) not used, sleep is intransitive

pilsen something to or for which one sleeps, rest, recuperation

pilsu something by or with which one sleeps, bed

8. voler lover

voli beloved

volen something to or for which one loves

volu something by or with which one loves

9. nawer/nawr someone or something who wants

nawi someone or something wanted

nawen/nawn something to or for which one wants, an intended recipient of a gift

nawu something by or with which one wants, desire

10. vaher someone who has, possessor

vahi something that is had or possessed

vahen something to or for which one has

vahu something by or with which one has

A system like this has two possible uses or results. In the first place, it could simply be used as a productive means of vocabulary building, such that isu, for example, would be used for optacle devices like spectacles or telescopes. On the other hand, this system could result in little need for the bulky relative clauses that English has to employ, unwieldy phrases like that by which (something) is seen, he who sees, that which I saw, and so forth. These could be reduced to one simple word each in such a language: isu, iser, isen.

Of course, there is no reason why both uses could not co-exist, but there could be ambiguity that results. This is not necessarily a bad thing, as it could be fruitful in producing puns and jokes, or clever poetic turns of phrase, but we might want some way of clarifying the meaning where necessary. Similarly, if we only had one of the two possible applications of this system, we would require other means to get the meanings that aren't covered. If we chose the first application of vocabulary building, we would probably have to have relative pronouns or particles to indicate the meanings covered by the second application. If we chose the second option of relativizing, we might have to have still more derivational suffixes to create words meaning lover and beloved, love, etc.

This is perhaps what interests me most of all about conlanging: devising various systems of grammar and testing their possible applications, and experimenting with the results of such systems, like lack of relativizers in the case of the example language here, or lack of adverbs in adjectives in my main conlang Paiodd. I find this stuff absolutely fascinating!