Siye [ne sʌⁿme] > Rural Ye [nɑⁿme] CAUS?
The Question of -mo
The initial verb-final aspectual suffixes are –me-, imperfective, and –ne-, perfective. To these could be attached the negative polarity suffix –u-, the emphatic suffix –a-, or the combination –a-u-. Thus the original PAM suffixes were imperfective positive normal –me, imperfective positive emphatic –mea, perfective negative normal –neu, and perfective negative emphatic –neau. The distinction between normal and emphatic shifted to one between irrealis and realis through overuse of the emphatic. Combination of the preceding suffixes with the clause-final interrogative suffix –mo produced the series –memo, -meamo, -meumo, -meaumo. One dialect (henceforth called dialect #1) reinterpreted the negative suffix –u- as the indicator of a negative question (clause-final) rather than the indicator of a negative polarity with the aspectual suffix (verb-final), which then is changed into a question (clause-final). Thus –meu-mo was reanalyzed as me-umo. Another dialect (henceforth dialect #2), developed this idea further by reanalyzing –mea-mo as –me-amo. Some conservative speakers of dialect #3, which had kept –mo, absorbed speakers of the former two dialects, and it was under their influence that the speakers of dialect #3 borrowed –umo and –amo as clause-final interrogative suffixes without regard to polarity, thereby creating dialect #4 with –meamo, -meaamo, -meuamo, -meauamo, and dialect #5 with –meumo, -meaumo, -meuumo, -meauumo. Vowel dominance reduced the dialect #4 series to –mamo, -mamo, -mewamo, -mawamo and the dialect #5 series to –mumo, -mumo, -mumo, -mumo. At the same time, the verb-final suffixes without the clause-final suffixes reduced in dialects #4 and #5 to –me, -ma, -mu, -mu. The homophony of negative realis and negative irrealis –mu was resolved by forming a new irrealis negative –meku from irrealis positive –me and alternative verb-final negative suffix –ku. This suffix –ku was a secondary suffix which attached to a primary suffix such as –me. Thus both dialects #4 and #5 had the series –me, -ma, -meku, -mu. The differing interrogative suffixes resulted in dialect #4 series –mamo, -mamo, -mekumo, -mumo and dialect #5 series –mumo, -mumo, -mekumo, -mumo. Dialect #5 then borrowed the –ku- of –mekumo and placed it in –mumo to produce –mukumo and thus a series –mumo, -mumo, -mekumo, -mukumo. It has never been clear why the positive interrogatives were never disambiguated in dialects #4 and #5. –mekumo could be broken down into –meku-mo from verb-final –meku and clause-final –mo, but –mukumo could not be broken down into a non-extant verb-final suffix –muku and clause-final –mo; nor could –mukumo be analyzed as verb-final –mu and clause-final –kumo, because –ku was neither a clause-final suffix such as –mo nor a primary suffix. The secondary suffix –ku had to be dependent on a preceding primary suffix. Thus dialect #5 innovated the clause-final negative realis suffix –ukumo as a parallel to the clause-final positive realis suffix –umo. This produced the dialect #5 series –mumo, -mumo, -mekumo, -mukumo, the familiar series in Standard Siye.
Are consonants from prefixes which are combined with a vowel-initial verb root aspirated? pe + a/tom produces pha or pa?
Evolution of PAM Suffix plus Interrogative Suffix
numo, nukumo, numo, nekumo > Ye nimo, ni:mo, nimo, ni:mo > Ye nimo, to nimo, ni:mo, to ni:mo
Siye stress is trochaic and dactylic. It is not particularly strong but it does make fine distinctions. Primary stress is on the verb root, secondary stress (if at all possible) on the directional suffix, and the PAM suffix (if at all possible) is unstressed. This last may be related to the modern Siye homophony of the frequently adjacent superdirectional suffix /na/ and the PAM suffix number 2 (perfective aspect, positive polarity, realis mood) /na/. Furthermore, a verb root or suffix may only take stress on its initial syllable; following syllables must be unstressed. Double syllables reduced to a single syllable retain the stress of any stressed syllable involved in the reduction. Stress on the final syllable is not permitted.
Prefixes are technically outside the stress system, thereby allowing the vowel of the first syllable of the verb root to receive primary stress (and the consonant of the first syllable of the verb root to be aspirated where appropriate), which allows (but does not permit – see converbal suffixes for the distinction) prefixes to be elided in daily fast speech if the context is clear. If there is one prefix or the surface form of the two prefixes reduces to one syllable, that syllable is unstressed. If the two prefixes remain bisyllabic, the origin of the prefixes as independent pronouns yields a strong secondary stress on the initial prefix. This stress is technically slightly stronger than the secondary stress on the directional suffix, since the directional suffix has been incorporated longer in the verb complex; often, however, the secondary stress on the prefix is flattened out to match the secondary stress on the directional suffix – especially if the backed subject has become the latest suffix that was formerly independent. In more careless speech, all of these secondary suffixes are reduced to equality.
Reading and Stress Analysis of A Siye Couplet
Ketekananukilo leneki elekelonana;
"Those things which I should not have done I have done;
Those things which I want to do I cannot do."
‘elekelonana’: e- is 3rd inanimate object prefix. –le- is the 1st subject prefix. –ke- is a perfective aspect suppletive root of ‘do’ or ‘die’, but here the object and subject prefix indicate that it is ‘do’, which is transitive, rather than ‘die’ , which is intransitive. If there is a causative suffix later on, reexamination may be necessary. –lo- is the plural grammatical number suffix. The aspect of the verb controls whether the grammatical number suffix refers to the subject or the object of the clause. In this case, the perfective aspect revealed by the use of the verb root –ke- indicates that the grammatical number suffix refers to the object e-. There is no causative suffix, nor any ‘converbal’ suffix. The directional suffix –na- could be considered part of a continuous root, -ke-na- ‘do’, although it is not mandatory here because the prefixes e-le- without a causative suffix indicate that the verb is inherently transitive rather than a valence-raised intransitive verb. The PAM suffix –na- indicates perfective aspect, positive polarity, and realis mood.
‘ketekananukilo’: ‘ketekananukilo’ is the plural form of the participial noun ‘ketekananaki’, which is a participial form of ‘elekelotekananu’ ‘I should not have done them’. e- is the 3rd inanimate object prefix, -le- is the 1st subject prefix, -ke-na- is the perfective transitive root ‘do’. –teka- is the ‘converbal’ suffix ‘would like to’ (with the irrealis mood) or ‘should’ (with the realis mood’). The PAM suffix –nu indicates realis mood, perfective aspect, and negative polarity, so –teka- is ‘should have not’ rather than ‘would not have liked to’. To form the participial stem, the prefixes and the grammatical number suffixes are removed; then, to form the passive inanimate stem, the aspect must be perfective (unless the ‘converbal’ suffix governs the imperfective) and the –ki suffix added. So ‘elekelotekananu’ becomes ‘ketekananuki’. The number suffix –lo is then added. Note that this –lo agrees with the –lo- in ‘elekelonana’. The absolutive case is zero-marked.
But the change from ‘elekelotekananu’ to ‘ketekananuki’ has eliminated reference to the subject. In neither word so far is there a reference to the grammatical number of the subject. One possibility is an explicit subject, but a different option is used here. The 1st person pronoun le- is suffixed with the Animate Instrumental case –neki. The singular number is zero-marked in Siye; if the pronoun were plural, the word would be ‘leloneki’.
‘elenupuyamnamu’: e- and –le- are the object and subject prefixes, respectively. –nu- is the imperfective suppletive root of transitive ‘do’, and formerly or archaically of ditransitive ‘give’, and the perfective suppletive root of transitive ‘copulate’. ‘give’ has yielded in finite non-participial forms to standard –numu-. –pu- is the singular grammatical number suffix. As yet, the aspect of the verb is not clear, and therefore neither is the subject or object reference of –pu-. The ‘converbal’ suffix –yam- indicates ‘be able to’. –yam- governs the imperfective aspect; therefore the verb is ‘do’ rather than ‘copulate’, and the grammatical number suffix refers to the subject rather than the object. The directional suffix –na- is an extension of the root ‘do’. In previous times, the use of –na- rather than –tu- distinguished –nu-na- ‘do’ from –nu-tu- ‘give’. The PAM suffix –mu has imperfective aspect (due to –yam-), realis mood, and negative polarity.
‘alenuputenamamelo’: ‘alenuputenanmamelo’ is the plural form of the nominalized relative verb ‘alenuputenamame’, which is the relativized form of ‘alenuputenama’. a- is the object-oriented object 3rd inanimate object prefix. –le- is the 1st subject prefix. –nu- is the imperfective transitive suppletive root ‘do’. –pu- is the singular grammatical number suffix and it applies to the subject because –nu- is imperfective . –te- is the ‘converbal’ suffix ‘to want’ rather than the suppletive ‘converbal’ suffix ‘to practice’ because the former governs the imperfective aspect while the latter governs the perfective aspect. The directional suffix –na- is an extension of the root ‘do’. The PAM suffix –ma- is positive polarity, imperfective aspect, and probably realis mood, although the following suffix –ame- could be obscuring irrealis mood PAM suffix –me-. The imperfective aspect here reconfirms that the ‘converbal’ suffix –te- is ‘to want’ and that the grammatical number suffix refers to the subject. The relative suffix –ame- transforms the verb into a relative form, but also obscures the mood of the PAM suffix. –lo is the plural number suffix. The absolutive case is zero-marked.
Ketekananukilo leneki elekelonana;
Elekelonana – The verb root is –ke-, the directional suffix is –na-, and the stressed prefix syllable is e-. Trochaic, trochaic, trochaic. Easy.
Leneki – This is a pronoun, but stress on the initial syllable produces a dactyl. Done
Ketekananukilo – This is a nominalized participle. The verb root is ke-, and the directional suffix is –na-. This creates a dactyl keteka-, and a four syllable –nanukilo. Note that the singular form, ketekananuki, consists of two dactyls and is therefore acceptable, but the stress assignment on the plural form is unfinished. –ki is a single syllable suffix and therefore can take the stress, creating a pattern of dactyl-dactyl-trochee. Also note that in ketekananukilo, the initial syllable of the suffix –teka- cannot take stress even though it is an initial syllable because ke- is stressed.
Elenupuyamnamu – The verb root is –nu-, the directional suffix is –na-, and the stressed prefix is e-. Trochee, dactyl, trochee. Resist the temptation to change trochaic –namu into dactylic –yamnamu. Although trochee-trochee-dactyl is a valid stress pattern, the directional suffix –na- has priority over the converbal suffix –yam-.
Alenuputenamamelo – This is a nominalized participle. The verb root is –nu-, the directional suffix is –na-, and the stressed prefix is a-. This produces a trochee ale-, a dactyl –nupute-, and –namamelo. Note that once again, if this had been singular, -namame would form a dactyl and stress assignment would be complete. The four syllable sequence –namamelo has only two potential stressed syllables: -na- and –ma-. –lo cannot receive stress because it is a final syllable, and –me- cannot receive stress because it is the second syllable of the relative suffix –ame-. –ma- here can receive stress because in case the –a- of –ma- is a contraction of the PAM suffix number one (imperfective aspect, positive polarity, realis aspect) and the relative suffix suffix –ame-. So secondary stress can move to –ma-, creating a dactyl –mamelo. This change, however, results in a single stressed syllable –na-. The stress then moves to –te-, a single syllable suffix and therefore eligible to receive stress, reducing the dactyl –nupute- to the trochee –nupu- and creating the trochee –tena-. The directional suffix –na- is the strongly preferred place for secondary stress, but it is not mandatory.