Proposal: Universal Extremal-Syllable Stress
|This page contains discussions of experimental/scientific/philosophical/logical aspects of Lojban that are|
non-official and not for everyday usage. You've been warned.
Krtisfranks proposes that Lojban be modified so that each word has mandatory stress on exactly its either first or♠ final syllable, excepting: cmevla (such that alternative stress is denoted in written form), words wherein such syllable has only "y" or a syllabic consonant as its nucleus, and other expressions which are demarcated by mandatory "." on either side. Notably, this would apply to cmavo as well as brivla.
♠ (This 'OR' statement is exclusive-definite: for affected polysyllabic words, the stress would be placed on the either first xor final syllable of said word, each; for monosyllabic words, the stress would be placed on the only syllable present (which is both the first and the final syllable) - in which case, the 'OR' becomes an 'AND').
Not all words are polysyllabic, so not all words can be stressed on their penult (because none exists). But all words, excepting the empty string/word, are at-least-monosyllabic and, thus, can be stressed on their either first xor final syllables (each of which is guaranteed to exist; again, note that equivalence of these syllables changes 'XOR' to 'AND').
Stress would be understood as meaning that a word boundary is 'imminent' in the sense that it either just happened or is about to happen. There is no wait for it.
Stress would be highly predictable. As it is now, a word can be modified and thereby lose its identifying stress pattern (for example: any CCVCVCV word derived from a CCVCV gismu changes the stress in the gismu from the gismu's first syllable to its last), or two words can be combined into a lujvo such that the same happens. Such would not necessarily happen (depending on the choice of proposal) if this proposal were adopted.
Morphology would be dramatically freed up. Many disallowed forms would be newly allowed because word boundaries would be less problematic to determine. A lot of the complexities of Lojban's morphological considerations (including things like slinku'i and tosmabru) are do to issues concerning uniquely identifying word boundaries; these issues would be alleviated and such words could be used without consternation or concern.
This would greatly reduce the complexity (especially for a human during a vocal conversation, in real time) in determining where word breaks occur and whether stress can be placed on a given syllable of a given cmavo in the particular context at hand. It also greatly simplifies the grammar rules regarding stress, which is especially beneficial while teaching newcomers.
- Presently, the grammar allows for free stress on cmavo except in certain circumstances; some have proposed that this exception may itself be deserving of exception (because it is an unnecessary restriction in some cases). However, keeping track of all of these exceptions can be difficult. This rule would eliminate such concerns.
It should(?) mean that some lujvo simplify in that no hyphen is required. Since word breaks would immediately precede or immediately follow the stressed syllable respectively (depending on which option of this proposal is chosen), rafsi strings which could/would be interpreted as containing cmavo (such as in the current grammar) would be interpreted as belonging to a single word, thereby yielding a lujvo. For example, "ro'ire'o" would be understood as a single word (either "rO'ire'o" xor "ro'ire'O" respectively, depending on proposal option), and so would be the lujvo "rokci zei renro"; the 2-cmavo string "ro'i re'o" would be distinguished from this as having two points of stress, rather than merely one, near word boundaries (namely, "rO'i rE'o" xor "ro'I re'O" respectively, similarly depending).
Words such that the otherwise affected syllable has only "y" or a syllabic consonant as its nucleus would be decomposed oddly. For example: "mlatu krtis" if rendered as "mlAtu krtis" or (worse) "mlAtu krtIs" would be interpreted as "mlAtukrtis" or "mlAtukr tIs" respectively. This might be especially bad for zi'evla with classifier rafsi/terms which are monosyllabic or disyllabic with only "y" or syllabic consonants present in at least the first syllable nucleus.