On Apparent “Systematic Suppletion” in Ksotre
by Angus Æ. Balderdash, Esq.
Julienne Autolycus, Ph.D.
Open Universe Open University
Jupiter City, Tvashtar Paterae, Io
As noted in Muddybanks (2007), not much is known of Ksotre in the English-speaking world, with Snodgrass (2001) being the previous (fairly low) height of academic interest. We appreciate Muddybanks’ willingness to reduce himself to a merely documentary linguistic approach in the face of the apparently unanalyzable and incomprehensible (but nonetheless intriguing) data.
Muddybanks closes his survey with this (obviously sincere) appeal:
In the opinion of the author, further research is required into the nature of the curious case system of Ksotre before anything definitive can be said. It is my hope, however, that the data presented above will whet the appetites of morphologists worldwide, and give birth to new research programmes centering on this little-studied, yet wonderfully rich and perplexing language.
Indeed, our interest has been piqued, our appetites whetted, and a (small) research program birthed. Our Universal Philology Department has been searching for a (more than) competent morphophonologist for a while now. We’ve been using the Ksotre “systematic suppletion” data as a test of (potential) morphophonologists for our (prestigious) department—the (original) goal was to see how long it would take a candidate to (finally) admit that they couldn’t figure out the data. Then (just recently) we got a candidate from Lithuania, whose (maternal) Grandmother is a (native) speaker of Ksotre! She submitted the following (stunning) analysis to us. We gave her (instant) tenure and (strongly) encouraged her to publish, despite her protestations that, in discussing the data with her Grandmother, she may have “gotten some of the (unimportant) details wrong.” The result is this (joint) article, combining the efforts of the Chair of the Universal Philology Department (Mssr. Balderdash) and its newest tenured professor (Dr. Autolycus).
There seem to be two main facts that (if unknown) confound the analysis of Ksotre, especially in the (most intriguing) Elative Dual:
First, Ksotre (rarely) has (underlyingly) so-called “strong” vowels. According to Daniel Jones’s (generally unknown) original description of the strong vowels, they (generally) cannot be pronounced, except by certain breeds of camels and the occasional (talented) flatulent penguin, and so are (always) transformed into “normal” vowels. We (following Jones) denote the strong vowels with bold: /i/ and /e/. The (most) important thing to know about the Elative Dual in Ksotre is that the Elative prefix is (underlyingly) actually /psi-/.
Second, a (potentially vast) number of deletion rules exist in Ksotre, which are motivated by a desire to avoid any of (vastly numerous) “taboo” syllables. Many of these taboo rules do not apply outside the Elative Dual (precisely) because other forms are (relatively) rare (see Muddybanks 2007, figure 6), and many a speaker isn’t (even) aware of the (rude) insult embedded in their speech until after it has slipped out. However, the Elative Dual, accounting for (almost) 50% of (inflected) noun forms, is streamedlined to avoid such faux pas—except (perhaps) when one wishes to be rude, then such taboo-deletion rules may (seemingly) be “forgotten”.
We don’t have any (theoretical) axes to grind, so we present our (stunning) findings about Ksotre phonology using sketches of simple (and familiar) pseudo-transformational rules. Careful details of rule ordering and additional (prohibitive) contexts, or (outright) conversion to optimality constraints are left as an (interesting) exercise for the (well-motivated) reader.
Strong Vowel Rules
Rules affecting (underlyingly) strong vowels are some of the most difficult to detect
(and to believe). But they have (enormous) explanatory power. More strong vowel rules will be presented (below) in the other sections.
strong V weakening — according to Daniel Jones’s (original) description, strong vowels cannot be pronounced, and so must be weakened. i → i, and e → e (as expected).
strong i spreading — strong /i/ can spread across one following morpheme boundary if all of the two or more vowels in the next morpheme are the same (and for (various) historical reasons, /o/ and /u/ count as the same).
e embiggening — e → e/i.__
strong i induced weak haplology — consecutive similar (but not identical) light syllables (C1V1C1V2) are reduced (to C1V2) if they follow strong i (at any distance).
strong i de-nasalization — n → ∅/i__
t → k/#__e
By far the most interesting rules in Ksotre are the “taboo” rules. Such deletions can (and do) erase morpheme boundaries. Here is a (random) sampling of such rules:
saw deletion — /saw/ is the (onomatopoetic) word for snoring, and is avoided in polite company and high registers; drop /saw/.
psili hamstringing — /psili/ is a (false) cognate with English “silly” and is to be avoided; drop /ps/, leaving /ili/.
nerine → nert taboo mitigation shift — /nerine/ is a slightly taboo word for excrement. /nert/ is an infantile (but phonologically similar) term for the same. This is (perhaps) akin to changing English scrappy to scacapy.
sivil deletion — /sivil/ is the (Ksotre) word for “syphilis”, and (as such) is taboo, and dropped.
post-sivil seppuku — if sivil deletion takes out (exactly) one morpheme, then all other phonemes commit (ritual) suicide, assisted by other phones. Vowels eliminate consonants, lower vowels eliminate higher vowels, until only one (the lowest) vowel remains.
e...w rule — /e ... w/ → /o ... ∅/; (yet) another taboo rule. The (underlying) root “ew” has a similar meaning (or lack thereof) to English “ewwww”, but is almost as taboo as the (infamous) English F-word. Thus, the two phonemes are not even allowed to remain (in the taboo order) in a word, even when separated by (considerable) distance.
Almost as interesting as the (more widespread) “taboo” rules are the (likely unique) morpheme mirroring rules, which are explained below.
morpheme mirror — reverses the order of morphemes in a (desperate) attempt to save the phonological derivation if one (or more) of the following conditions is satisfied:
unpronounceability — if, through other (normal) processes, a morpheme becomes unpronounceable, it is reversed.
anti-analogy — if an intermediate form is (very) similar to a (common) final form, the center-most morpheme is mirrored.
tilted morpheme mirror semi-assimilation — a morpheme will be flipped (vertically) if, in a two-morpheme word, one morpheme is flipped (horizontally). (Interestingly, the (orthographic) regularity of this rule hints at (significantly) less-than-expected arbitrariness in IPA symbols—whether this is indicative of iconicity (of the symbols) or hard-wiring (of the symbol makers) is an open question.)
mirror breaking — to block (commonly) unproductive mirroring, /psr/ defensively simplifies to /sr/.
strong e epenthesis — if, after morpheme mirror, a morpheme is (still) unpronounceable, strong e is inserted to separate unacceptable clusters
Historical Holdover Rules
These are rules that are best explained by looking at the ((surprisingly) well-documented) historical changes that have happened in Ksotre.
orthographic breaking of mono-morphemic palindromes — if an open-class morpheme with a CVC shape is also a palindrome, then, if it is possible to erase (some portion of) the printed representation of the initial and final phonemes (which are the same, of course) such that the result is readily interpreted as a representation of valid (non-palindromic) Ksotre orthography, then do so. Thus, /mim/ → /rnirn/, /dod/ → /clocl/, but not /wew/ → /vvevv/ because the result is (still) palindromic. Many Ksotre philologists blame this (inelegant) change on the (generally poor) quality of OCR software in use throughout Ksotre-speaking areas during the 1980s.
“Sartre”/“Sit” alternation — a (very) rude but (very) influential response to Sartre’s La Nausée was written in the 1940s by a (still) unknown author using the pseudonym “Jean-Paul Sit”. Interestingly, “sit” is an elative dual for a (scatological) term we need not define here. In all the (trendy) cafes across the land, every mention of “Sartre” was replaced with “Sit” (Part 1). In order to highlight the joke, hidden away in such a short syllable, any (inflectional) morphemes attached to “Sartre”-cum-”Sit” were dropped (Part 2). As the combination /artre/ is (vanishingly) rare in Ksotre, the /artre/ → /it/ transformation, which began as a joke, became grammaticalized. This (initially) pro-taboo rule is an interesting (sociolinguistic) exception to the general trend in Ksotre.
harmonic reduplication — Later Paleo-Medieval Ksotre had i. productive reduplication for emphasis, ii. vowel harmony, iii. consonant harmony, and iv. a very strong poetic oral tradition. A (rare) holdover from those days is an interesting (but obvious) intersection of these processes, which resulted in the syllable after reduplicated /tr/ being reduplicated as well. Once the phonotactics of the language changed and /trtr/ was no longer an acceptable cluster, the rule spread to all instances of /rtr/. Note that /aw/ reduplicates to /avaw/ because of interference from the (prestigious) /v/ ↔ /w/ dialect. This rule, always uncommon, rarely applies nowadays because it is (so often) bled by the Sartre/Sit rule.
Additional Phonological Rules
These are the (fairly boring) rules that do the day-to-day heavy phonological lifting in Ksotre.
i → ∅/__.V
non-phoneme V — vowels not in the (normal) phonological inventory of Ksotre are converted to /i/.
prestigious /v/ ↔ /w/ swap — imitate (high-prestige) dialect variation if both /v/ and /w/ are present, and swap the two.
radical epenthetic liquid dissimilation — hypothesized for the (future) development of English by Jay Trones (1993), and labeled “as of yet unseen”, this transformation is (in fact) well-attested in Ksotre.
triple i rule — /iCiCi/ → /ilili/. This rule—like the “CaReFuL” pronunciation rule (sometimes) taught to English-speaking second-language learners of French, or the “Ruki” sound law in the Satem languages—has no (known) motivation. This is a (surprisingly) uncommon situation in Ksotre.
Below we present some (simple) examples, taken from the data presented in Muddybanks (2007), to illustrate these (perfectly natural) phonological principles.
|ps.ertrav.aw → wit|
| || psi.ertrav.aw|
| psi.vartre.aw|| ||morpheme mirror|
(see /psertravaw/, elative dual of /mim/)
| psi.wartre.av||prestigious /v/ ↔ /w/ swap|
| psi.wit.av||Sartre/Sit alternation, part 1|
| wit||Sartre/Sit alternation, part 2|
|psi.topos.aw → ilili|
| psi.tipis.aw||strong i spreading|
| psi.tipis.aw||strong i weakening|
| psi.lilis.aw||triple i rule|
| i.lilis.aw||psili rule|
| i.lili||saw rule|
|ps.ovult.aw → kepim|
| psi.ivilt.aw||strong i spreading|
| psi.ivilt.aw||strong i weakening|
| ps.ivilt.aw||i → ∅/__.V|
| pt.aw||sivil deletion|
| tp.aw||morpheme mirror|
(/pt/ is not acceptable in Ksotre)
| tp.ɐm||tilted morpheme mirror|
| tp.im||non-phoneme V|
| tep.im||strong e epenthesis|
| kep.im||t → k/#__e|
| kep.im||strong e weakening|
|ps.elalit.aw → srostra|
| || psi.elalit.aw|
| psi.elit.aw|| ||strong i induced weak haplology|
| psi.elit.aw||e embiggening|
| psi.elit.aw||strong i weakening|
| ps.elit.aw||i → ∅/__.V|
| psr.elitr.aw||radical epenthetic liquid dissimilation|
| sr.elitr.aw||/psr/ → /sr/ (mirror breaking)|
| sr.olitr.a||ew rule|
| sr.oʃtr.a||abnormal excessive palatalization|
| sr.ostr.a||ʃ → s|
|psi.saval.aw → a|
| psi.sivil.aw||strong i spreading|
| psi.sivil.aw||strong i weakening|
| psi.aw||sivil taboo|
| a||post-sivil seppuku|
|psi.mim.aw → psertravaw|
| psi.rninrn.aw||orthographic breaking of|
| psi.nrinr.aw||morpheme mirror|
| psi.neriner.aw||strong e epenthesis|
| psi.neriner.aw||strong e weakening|
| psi.nertr.aw||nerine → nert taboo mitigation shift|
| psi.ertr.aw||strong i de-nasalization|
| psi.ertr.aw||strong i weakening|
| psi.ertr.avaw||harmonic reduplication|
| ps.ertr.avaw||i → ∅/__.V|
Similar (perfectly natural) analyses explain these three additional forms (also from Muddybanks), and many others:
psi.tsuip.aw → orso
psi.lefked.aw → hontordarus
ps.ankuf.aw → deskovurtiva
Anderson, Stephen R., 1981. “Why phonology isn’t ‘natural’,” Linguistic Inquiry 12.493-539.
Bach, Emmon, and Robert Harms, 1972. “How do languages get crazy rules?” Linguistics change and generative theory, ed. by Robert Stockwell and Ronald Macaulay, 1-21.
Jones, Daniel, 2002. Daniel Jones: Collected Works, Vols. 1-8, ed. B. Collins and I.M. Mees.
Muddybanks, Lawrence R., 2007. “Systematic Suppletion: An Investigation of Ksotre Case Marking,” Speculative Grammarian, CLII.2.
“Sit, Jean-Paul”, 1940. “Nauséeuzaw,” Journal Of Ksotre Literature and Philosophy Studies, IV.7.
Snodgrass, Quentin P., 2001. Ksotre Phonology: It has one.
Trones, Jay, 1993. “The Boustrophedon-Plummerfeld Hypothesis,” Speculative Grammarian, CXLVII.2.