Dissimilation¶
A phonological change or alternation in which one sound becomes less like a similar sound in a specified context.
Core Idea¶
Dissimilation is a phonological change or alternation in which one sound becomes less like another similar sound in a specified context. Its analysis identifies the pair, affected feature, target sound and domain; it does not require that the sounds be adjacent or that the first one always change. Nor does it establish repetition avoidance as a universal psychological cause.[ref-f3d58fda6f55][ref-efcf03918e41]
Latin -ālis/-āris adjective formation and Greek/Sanskrit aspirate dissimilation show the common relation without sharing one sound law. A Latin stem /l/ can favor rhotic /r/ in the later suffix; Grassmann's historical pattern generally removes aspiration from an earlier consonant when two relevant aspirates are present. Feature, direction and language scope remain distinct.[ref-f3d58fda6f55][ref-efcf03918e41]
Scope of Application¶
For Latin adjective formation, Müller documents the -ālis baseline, stem-lateral-conditioned -āris, an intervening-/r/ condition and lexical exceptions. Mīles with stem mīlit- yields mīlitāris in the cited examples. Suzuki also illustrates sol-alis → solaris and lun-alis → lunaris. This is a bounded suffixal pattern, not a rule that any two /l/ sounds anywhere in Latin must change.[ref-f3d58fda6f55][ref-f2323b22b8dd]
Grassmann's original Greek/Sanskrit account instead considers relevant pairs of aspirated consonants in historical root configurations; one, generally the first, loses aspiration. The Latin and aspirate cases each have a similar pair and a less-similar outcome, but one cannot move Latin's suffix, target direction or exceptions into Greek by analogy.[^ref-efcf03918e41]
Clarity¶
Dissimilation names an observed conditioned divergence, not necessarily its cause. “Speakers dislike repetition,” perceptual hypercorrection and other explanations require their own evidence. It differs from assimilation, which increases the relevant similarity, and from an unrelated sound change, which lacks a demonstrated similar trigger.[^ref-f3d58fda6f55]
It also differs from live Phonological Process, whose identity focuses on child speech-production rules, developmental suppression and clinical use. Historical Latin/Greek sound changes need none of those developmental roles. A surface -āris ending alone is not proof: the environment, baseline and known exceptions matter.[ref-f3d58fda6f55][ref-f2323b22b8dd]
Manages Complexity¶
The Latin suffix alternation compresses many adjective forms into a relation between stem liquids and suffix choice, while the Greek/Sanskrit case shows that the same cross-case question can be asked of aspiration. That makes patterns comparable without treating every word as isolated. The compression has a cost: intervention, exceptions, feature choice and history have to remain attached to concrete predictions.[ref-f3d58fda6f55][ref-efcf03918e41]
Abstract Reasoning¶
Start with two sound units that share a feature in a defined domain. Ask whether the outcome changes one unit so the pair shares less of that feature. Latin stem /l/ plus adjectival suffixal /l/ can yield suffixal /r/ in -āris; the changed sound is later. Grassmann's configuration has two aspirates and generally deaspirates the earlier one. In each case the relation is sound similarity → conditioned target → reduced similarity, followed by a check of language-specific scope, interveners and exceptions.[ref-f3d58fda6f55][ref-efcf03918e41][^ref-f2323b22b8dd]
Knowledge Transfer¶
The Latin-to-Grassmann transfer is literal only at the level of a context-linked similarity-reduction relation. Liquid identity and suffix morphology are not inherited by the aspiration case, nor does one target direction become universal. To identify a fresh case, supply comparative forms, the relevant similar feature, a conditioned divergent output and the bounded language domain.[ref-f3d58fda6f55][ref-efcf03918e41]
The broader idea of context-driven divergence may merit future prime study, but this evidence shows phonological reuse, not a substrate-independent prime. No checked live genus captures both historical sound-change cases without adding unrelated child-developmental or disciplinary requirements; this draft stages an unparented domain-specific node.
[^ref-f3d58fda6f55]: Daniela Müller, “Liquid dissimilation with a special regard to Latin” (2013), author-hosted manuscript, §2 examples (3)–(5) and Table 1, §§3 and 7 on competing explanations. https://www.phonetik.uni-muenchen.de/personen/assoziierte_wissenschaftler/mueller_daniela/publications/mueller2013liquiddissimilation.pdf [^ref-efcf03918e41]: Hermann Grassmann, “Concerning the Aspirates and Their Simultaneous Presence in the Initial and Final of Roots,” 1863 original in Zeitschrift für vergleichende Sprachforschung 12(2), 81–138; English translation hosted by the University of Texas Linguistics Research Center in A Reader in Nineteenth Century Historical Indo-European Linguistics, Chapter 10. https://lrc.la.utexas.edu/books/reader/10-h-grassmann [^ref-f2323b22b8dd]: Keiichiro Suzuki, A Typological Investigation of Dissimilation, University of Arizona PhD dissertation (1998), Rutgers Optimality Archive copy, PDF pp. 101–102, examples (46)–(47). https://roa.rutgers.edu/files/281-0998/roa-281-suzuki-1.pdf
Neighborhood in Abstraction Space¶
Dissimilation sits in a sparse region of the domain-specific corpus (61st percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.
Family — Language, Mind & Meaning-Making (57 abstractions)
Nearest neighbors
- Folk Etymology — 0.86
- Rhyme — 0.85
- Perception of English /r/ and /l/ by Japanese speakers — 0.85
- Teaching prosody — 0.84
- Empty category principle — 0.84
Computed from structural-signature embeddings · 2026-10-08