Community hub
Recent from talks
Contribute something
Nothing was collected or created yet.
L-vocalization
View on Wikipedia
 | This article needs additional citations for verification. (September 2016) |
| Sound change and alternation |
|---|
| Fortition |
| Dissimilation |
L-vocalization, in linguistics, is a process by which a lateral approximant sound such as [l], or, perhaps more often, velarized [ɫ], is replaced by a vowel or a semivowel.
Types
[edit]There are two types of l-vocalization:
- A labiovelar approximant, velar approximant, or back vowel: [ɫ] > [w] or [ɰ] > [u] or [ɯ]
- A front vowel or palatal approximant: [l] > [j] > [i]
West Germanic languages
[edit]Examples of L-vocalization can be found in many West Germanic languages, including English, Scots, Dutch, and some German dialects.
Early Modern English
[edit] | This section needs additional citations for verification. (February 2014) |
L-vocalization has occurred, since Early Modern English, in certain -al- and -ol- sequences before coronal or velar consonants, or at the end of a word or morpheme. In those sequences, /al/ became /awl/ and diphthonged to /ɑul/, while /ɔl/ became /ɔwl/ and diphthonged to /ɔul/.[1]
At the end of a word or morpheme, it produced all, ball, call, control, droll, extol, fall, gall, hall, knoll, mall, pall, poll, roll, scroll, small, squall, stall, stroll, swollen, tall, thrall, toll, troll and wall. The word shall did not follow this trend, and remains /ˈʃæl/ today.
Before coronal consonants, it produced Alderney, alter, bald, balderdash, bold, cold, false, falter, fold, gold, halt, hold, malt, molten, mould/mold, old, palsy, salt, shoulder (earlier sholder), smolder, told, wald, Walter and wold (in the sense of "tract of land"). As with shall, the word shalt did not follow the trend and remains /ˈʃælt/ today.
Before /k/, it produced balk, Balkans, caulk/calk, chalk, Dundalk, falcon, folk, Polk, stalk, talk, walk and yolk.
Words like fault and vault did not undergo L-vocalization but rather L-restoration. They had previously been L-vocalized independently in Old French and lacked the /l/ in Middle English but had it restored by Early Modern English. The word falcon existed simultaneously as homonyms fauco(u)n and falcon in Middle English. The word moult/molt never originally had /l/ to begin with and instead derived from Middle English mout and related etymologically to mutate; the /l/ joined the word intrusively.
L-vocalization established a pattern that would influence the spelling pronunciations of some relatively more recent loanwords like Balt, Malta, polder, waltz and Yalta. It also influenced English spelling reform efforts, explaining the American English mold and molt as opposed to the traditional mould and moult.
However, certain words of more recent origin or coining do not exhibit the change and retain short vowels, including Al, alcohol, bal, Cal, calcium, doll, gal, Hal, mal-, Moll, pal, Poll, Sal, talc, and Val.
While in most circumstances L-vocalization stopped there, it continued in -alk and -olk words, with the /l/ disappearing entirely in most accents (with the notable exception of Hiberno-English). The change caused /ɑulk/ to become /ɑuk/, and /ɔulk/ to become /ɔuk/. Even outside Ireland, some of these words have more than one pronunciation that retains the /l/ sound, especially in American English where spelling pronunciations caused partial or full reversal of L-vocalization in a handful of cases:
- caulk/calk can be /ˈkɔːlk/ or /ˈkɔːk/.
- falcon can be /ˈfælkən/, /ˈfɔːlkən/ or /ˈfɔːkən/.
- yolk can be /ˈjoʊlk/ or /ˈjoʊk/; yoke as /ˈjoʊk/ is only conditionally homophonous.
The Great Vowel Shift changed L-vocalized diphthongs to their present pronunciations, with /ɑu/ becoming the monophthong /ɔː/, and /ɔu/ raising to /ou/.
The loss of /l/ in words spelt with -alf, -alm, -alve and -olm did not involve L-vocalization in the same sense, but rather the elision of the consonant and usually the compensatory lengthening of the vowel.
Modern English
[edit]More extensive L-vocalization is a notable feature of certain dialects of English, including Cockney, Estuary English, New York English, New Zealand English, Pittsburgh English, Philadelphia English and Australian English, in which an /l/ sound occurring at the end of a word (but usually not when the next word begins with a vowel and is pronounced without a pause) or before a consonant is pronounced as some sort of close back vocoid: [w], [o] or [ʊ]. The resulting sound may not always be rounded. The precise phonetic quality varies. It can be heard occasionally in the dialect of the English East Midlands, where words ending in -old can be pronounced /oʊd/. K. M. Petyt (1985) noted this feature in the traditional dialect of West Yorkshire but said it has died out.[2] However, in recent decades, l-vocalization has been spreading outwards from London and the southeast;[3][4] John C. Wells argued that it is probable that it will become the standard pronunciation in England over the next one hundred years,[5] which Petyt criticized in a book review.[6]
For some speakers of the General American accent, /l/ before /f v/ (sometimes also before /s z/) may be pronounced as [ɤ̯].[7]
In Cockney, Estuary English, New Zealand English and Australian English, l-vocalization can be accompanied by phonemic mergers of vowels before the vocalized /l/, so that real, reel and rill, which are distinct in most dialects of English, are homophones as [ɹɪw].
Graham Shorrocks noted extensive L-vocalization in the dialect of Bolton, Greater Manchester, and commented, "many, perhaps, associate such a quality more with Southern dialects, than with Lancashire/Greater Manchester."[8]
In the accent of Bristol, syllabic /l/ can be vocalized to /o/, resulting in pronunciations like /ˈbɒto/ (for bottle). By hypercorrection, however, some words originally ending in /o/ were given an /l/: the original name of Bristol was Bristow, but this has been altered by hypercorrection to Bristol.[9] In Plymouth L-vocalization is also found, but without turning into the Bristol L afterwards.
African-American English dialects may have L-vocalization as well. However, in these dialects, it may be omitted altogether: fool becomes [fuː]. Some English speakers from San Francisco, particularly those of Asian ancestry, also vocalize or omit /l/.[10]
German
[edit]In colloquial varieties of modern standard German, including the northern Missingsch, there is a moderate tendency to vocalize coda /l/ into /ɪ̯/, especially in casual speech. This is most commonly found before /ç/ in words like welche ("which") or solche ("such"), which merges with Seuche ("disease"). To a lesser degree, the same may also occur before other dorsal and labial consonants.
A similar but far more regular development exists in many dialects of Austro-Bavarian, including Munich and Vienna. Here, etymological /l/ in the coda is vocalized into i or y in all cases. For example, Standard German viel ("much") corresponds to vui in Munich, vey in parts of the Southeast, vii North, West and East of Salzburg throughout the state of Salzburg, into Bavaria and into Upper Austria,[11] especially well pronounced on the German side of the border, and vü in Viennese.
In most varieties of the Bernese dialect of Swiss German, historical /l/ in coda position has become [w] and historical /lː/ (only occurring intervocalically) has become [wː], whereas intervocalic /l/ persists. The absence of vocalization was one of the distinctive features of the now-uncommon upper-class variety. It is still missing from dialects spoken in the Bernese Highlands and, historically, in the Schwarzenburg area. For example, the Bernese German name of the city of Biel is pronounced [ˈb̥iə̯w].
This type of vocalization of /l/, such as [sɑwts] for Salz, is recently spreading into many Western Swiss German dialects, centred around Emmental.
Middle Scots
[edit]In early 15th century Middle Scots /al/ (except, usually, intervocalically and before /d/), /ol/ and often /ul/ changed to /au/, /ou/ and /uː/. For example, all changed to aw, colt to cowt, ful to fou (full) and the rare exception hald to haud (hold).
Middle Dutch
[edit]![[icon]](https://en.wikipedia.org/wiki/File:Wiki_letter_w_cropped.svg){kind=small} | This section needs expansion. You can help by adding to it. (September 2012) |
In early Middle Dutch, /ul/, /ol/ and /al/ merged and vocalized to /ou/ before a dental consonant (/d/ or /t/):
- schouder "shoulder" < schulder
- oud "old" < ald
- hout "wood" < holt
- Wouter, a name < Walter
The combination /yl/, which was derived from /ol/ or /ul/ through umlaut, was not affected by the change, which resulted in alternations that still survive in modern Dutch:
- goud "gold", but gulden "golden"
- schout "sheriff", but schuld "guilt, debt"
- zouden "would" < zolden, past tense of zullen "to will, shall"
Ablaut variations of the same root also caused alternations, with some forms preserving the /l/ and others losing it:
- houden "to hold", past tense hield
- wouden "wanted" < wolden, past tense of willen "to want"
Analogy has caused it to be restored in some cases, however:
- wilden reformed next to older wouden
- gelden "to apply", past tense golden, earlier gouden
Modern Dutch
[edit]Many speakers of the northern accents of Dutch realize /l/ in the syllable coda as a strongly pharyngealized vowel [ɤ̯ˤ].[12]
L-vocalization increased significantly from 1957, especially among women and people from Holland and Utrecht areas.[13]
In some dialects, instead of vocalization, it is more common to pronounce a clearer [ə] (a Svarabhakti vocal) after a vowel followed by a [l] or [r]: melk (milk) becomes [mel·ək].
Romance languages
[edit] |
French
[edit]In pre-Modern French, [l] vocalized to [u] in certain positions:
- between a vowel and a consonant, as in Vulgar Latin caldu(m) "warm, hot" > Old French chaud /tʃaut/
- after a vowel at the end of a word, as in Vulgar Latin bellu(m) > Old French bel > Old French beau /be̯au̯/ "beautiful" (masculine singular; compare the feminine belle /bɛlə/, in which the l occurred between vowels and did not vocalize)
By another sound change, diphthongs resulting from L-vocalization were simplified to monophthongs:
- Modern French chaud [ʃo]
- Modern French beau [bo] (belle [bɛl])
Italo-Romance languages
[edit]In early Italian, /l/ vocalized between a preceding consonant and a following vowel to /j/: Latin florem > Italian fiore, Latin clavem > Italian chiave.
Neapolitan shows a pattern similar to French, as [l] is vocalized, especially after [a]. For example, vulgar Latin altu > àutə; alter > àutə; calza > cauzétta (with diminutive suffix). In many areas the vocalized [l] has evolved further into a syllabic [v], thus àvətə, cavəzetta.
Ibero-Romance languages
[edit]West Iberian languages such as Spanish and Portuguese had similar changes to those of French, but they were less common: Latin alter became autro and later otro (Spanish) or outro (Portuguese), while caldus remained caldo, and there were also some less regular shifts, like vultur to buitre (Spanish) or abutre (Portuguese).
In Portuguese, historical [ɫ] (/l/ in the syllable coda) has become [u̯ ~ ʊ̯] for most Brazilian dialects, and it is common in rural communities of Alto Minho and Madeira. For those dialects, the words mau (adjective, "bad") and mal (adverb, "poorly", "badly") are homophones and both pronounced as [ˈmaw]~[ˈmaʊ], while standard European Portuguese prescribes [ˈmaɫ]. The pair is distinguished only by the antonyms (bom [ˈbõ]~[ˈbõw] and bem [ˈbẽj]).
Slavic languages
[edit]South Slavic languages
[edit]In Standard Serbo-Croatian, historical /l/ in coda position has become /o/ and is now so spelled at all times in Serbian and most often in Croatian. For example, the native name of Belgrade is Beograd (Croatia also has a town of Biograd). However, in some final positions and in nouns only, Croatian keeps the /l/ by analogy with other forms: stol, vol, sol vs. Serbian sto, vo, so (meaning "table", "ox" and "salt" respectively). This does not apply to adjectives (topao) or past participles of verbs (stigao), which are the same in Standard Croatian as in Standard Serbian.
In Slovene, historical coda /l/ is still spelled as l but almost always pronounced as [w].
In Bulgarian, the phoneme /ɫ/ is pronounced as a labio-velar approximant [w] in all positions in certain urban dialects, particularly among young people. For example, words such as лодка, малка are pronounced [ˈwɔtkɐ], [ˈmawkɐ]. This feature is also associated with certain traditional dialects, mostly around Pernik, though it is thought that it evolved independently in the rest of the country. Still, it is more widespread in western dialects.
Polish and Sorbian
[edit]In Polish and Sorbian languages, almost all historical /ɫ/ have become /w/, even in word-initial and inter-vocalic positions. For example, mały ("small" in both Polish and Sorbian) is pronounced by most speakers as [ˈmawɨ] (compare Russian малый [ˈmalɨj]). The [w] pronunciation, called wałczenie in Polish, dates back to the 16th century, first appearing among the lower classes. It was considered an uncultured accent until the mid-20th century, when the stigma gradually began to fade. As of the 21st century, [ɫ] is still used by some speakers of eastern Polish dialects, especially in Belarus and Lithuania, as well as in Polish-Czech and Polish-Slovak contact dialects in southern Poland.[14]
Ukrainian and Belarusian
[edit]In Ukrainian and Belarusian, in the syllable coda, historical /ɫ/ has become [w] (written ⟨в⟩ in Ukrainian and ⟨ў⟩ in Belarusian, now commonly analyzed as coda allophone of /ʋ/–/v/). For example, the Ukrainian and Belarusian word for "wolf" is вовк [ʋɔwk] and воўк [vowk] as opposed to Russian вoлк [voɫk]. The same happens in the past tense of verbs: Russian дал [daɫ], Ukrainian дав [daw], Belarusian даў [daw] "gave". The /ɫ/ is kept at the end of nouns (Russian and Belarusian стoл [stoɫ], Ukrainian стіл [stiɫ] "table") and before suffixes (before historical ⟨ъ⟩ in the word middle): Russian, Ukrainian, and Belarusian палка [ˈpaɫka] "stick".
Uralic languages
[edit]Proto-Uralic *l was vocalized to *j in several positions in the Proto-Samoyed language. Several modern Uralic languages also exhibit l-vocalization:
- In Hungarian, former palatal lateral *ʎ (still written by a separate grapheme ly) has become a semivowel /j/.
- A similar phenomenon exists in Swedish, where initial */lj/ (written by lj) has also became /j/ (this does not apply to Swedish spoken in Finland).
- Most Zyrian dialects of Komi vocalize syllable-final /l/ in various ways, which may result in [v], [u], or vowel length.
- Veps also vocalizes original syllable-final *l to /u/.
See also
[edit]References
[edit]- Labov, William, Sharon Ash, and Charles Boberg. 2006. The Atlas of North American English. Berlin: Mouton de Gruyter. ISBN 3-11-016746-8.
- ^ Jesperson, Otto (1954). A Modern English Grammar vol. 1. London: Bradford & Dickens. pp. 289–297.
- ^ KM Petyt, Dialect & Accent in Industrial West Yorkshire, John Benjamins Publishing Company, page 219
- ^ Asher, R.E., Simpson, J.M.Y. (1993). The Encyclopedia of Language and Linguistics. Pergamon. p. 4043. ISBN 978-0080359434
- ^ Kortmann, Bernd et al. (2004). A Handbook of Varieties of English. Mouton de Gruyter. p. 196. ISBN 978-3110175325.
- ^ Wells, John C. (1982). Accents of English. Cambridge University Press. p. 259. ISBN 978-0521297196
- ^ Petyt, KM (1982). "Reviews: JC Wells: Accents of English". Journal of the International Phonetic Association. 12 (2). Cambridge: 104–112. doi:10.1017/S0025100300002516. S2CID 146349564.
- ^ Rogers, Henry (2000), The Sounds of Language: An Introduction to Phonetics, Essex: Pearson Education Limited, pp. 120–121, ISBN 978-0-582-38182-7
- ^ Shorrocks, Graham (1999). A Grammar of the Dialect of the Bolton Area. Pt. 2: Morphology and syntax. Bamberger Beiträge zur englischen Sprachwissenschaft; Bd. 42. Frankfurt am Main: Peter Lang. p. 255. ISBN 3-631-34661-1. (based on the author's thesis (Ph. D.)--University of Sheffield, 1981)
- ^ Harper, Douglas. "Bristol". Online Etymology Dictionary.
- ^ L Hall-Lew and R L Starr, Beyond the 2nd generation: English use among Chinese Americans in the San Francisco Bay Area, English Today: The International Review of the English Language, vol. 26, issue 3, pp. 12-19. [1]
- ^ "Dialektlandschaft Salzkammergut". www.sprachatlas.at. Retrieved 2025-04-10.
- ^ Collins, Beverley; Mees, Inger M. (2003), The Phonetics of English and Dutch, Fifth Revised Edition, E.J. Brill, pp. 197 and 287, ISBN 9004103406
- ^ van Reenen, Pieter (2000), De Vocalisering van de /l/ in het Standaard Nederlands (PDF)
- ^ Leksykon terminów i pojęć dialektologicznych : Wałczenie
External links
[edit]- Transcribing Estuary English, by J. C. Wells - discusses the phonetics of l-vocalization in Estuary English and Cockney.
L-vocalization
View on GrokipediaDefinition and Phonological Processes
Core Definition
L-vocalization is a phonological process in which a lateral approximant , or more commonly its velarized variant [ɫ], occurring in syllable coda position, is replaced by a vowel or semivowel, such as , , , , or .[4] This change typically affects the dark [ɫ], which features tongue retraction toward the velum, transforming it into a more vowel-like segment while preserving the original place of articulation to varying degrees.[1] The process is conditioned by the syllable rhyme structure, leaving onset unaffected.[4] Historically, L-vocalization first appears in records of Indo-European languages during the Middle Ages, with evidence in Old French from before the 10th century and generalization in the 11th and 12th centuries.[5] In early Scots, a Germanic variety, it emerges in the early 15th century, particularly after back vowels in coda position, as seen in forms like *healf > hawff.[6] Cross-linguistically, it manifests as a natural sound change that increases sonority by converting a consonant into a higher-sonority vocoid, observed in languages like Serbo-Croatian, Polish, Catalan, and Mehri.[4] Unlike complete lateral deletion, where the segment is entirely lost (e.g., American English help [hɛp] in certain varieties), L-vocalization maintains the phonetic material as a glide or vowel, avoiding total effacement.[5] This preservation distinguishes it from effacement processes in historical Romance changes, where pre-consonantal might disappear without trace after certain vowels.[5] For instance, in de-lateralization pathways, vocalization involves reduction of the apical gesture to produce a vocoid, rather than outright deletion.[7] Representative examples illustrate the process: in English dialects like London English, milk is realized as [mɪwk], reflecting a labiovelar shift.[8]Phonetic Variants
In the process of L-vocalization, the velarized lateral approximant [ɫ], typically occurring in syllable coda positions, undergoes articulatory simplification where the coronal (tongue-tip) contact with the alveolar ridge is lost or weakened, while the dorsal (back-of-tongue) retraction and associated lip rounding persist or intensify. This results in a transition to a back rounded glide or a vowel such as [ʊ], , or , as the gesture reduces from a complex lateral approximant to a single vocoid-like articulation.[1][9] Acoustically, this change is marked by formant transitions, particularly a lowering of the second formant (F2), which drops from approximately 1500 Hz in a clear to around 800 Hz in the resulting or back vowel, reflecting the increased tongue body retraction and rounding. Spectrographic analyses of English dialects, such as those in Southern British varieties, illustrate this as a gradual F2 descent in coda [ɫ], often blending seamlessly into the preceding vowel's trajectory and creating a diphthongal effect.[10][1] Variant realizations range from partial labialization, where [ɫ] becomes a non-lateral off-glide [o̯] retaining some consonantal quality, to full vocalization as a syllabic or non-syllabic vowel , depending on prosodic and segmental context. The preceding vowel influences the outcome; for instance, a high front vowel like /ɪ/ in sequences such as /ɪl/ may yield [ɪw], preserving front-back contrast through the glide, while back or low vowels promote more centralized back vowel results.[1][9] Cross-linguistically, dark [ɫ] exhibits a stronger tendency to vocalize than clear due to its inherent dorsal component, which aligns with the sonority hierarchy where liquids like [ɫ] have higher sonority than obstruents, facilitating their promotion to vocalic status in coda environments across languages such as English dialects and certain Romance varieties.[1]Triggering Environments
L-vocalization primarily affects the lateral approximant /l/ when it occurs in the syllable coda position, following a vowel, either before another consonant (V-l-C) or at a word boundary (V-l-#). This process targets post-vocalic contexts where /l/ is typically realized as a dark [ɫ] with a raised tongue dorsum, facilitating its rearticulation as a vowel or glide. In contrast, /l/ in syllable onset positions (e.g., #l-V or C-l-V) remains unchanged, preserving its clear lateral quality.[11][1][12] Adjacency to certain segments significantly influences the likelihood of vocalization. It is more frequent following back or low vowels (e.g., /ɔl/ > [ɔw]), where the dorsal gesture of dark /l/ assimilates readily to the preceding vowel's backness, enhancing the transition to a rounded vocoid. Conversely, front vowels disfavor vocalization due to greater articulatory conflict with the coronal-lateral properties of /l/. Following coronals can block or reduce vocalization rates (e.g., around 31% after coronals vs. 50% after labials), as the lateral airflow is maintained for plosive release.[3][12][1] Prosodic structure plays a key role in conditioning vocalization, with higher rates observed in word-final positions or before pauses, where the lack of a following vowel allows the dorsal gesture to dominate without interference. Syllabic /l/ (e.g., in "bottle") often shows lower vocalization (around 36-58%), as its nuclear status reinforces sonority but resists full delateralization. Foot boundaries may limit spread by preserving /l/ integrity in stressed or prosodically prominent contexts, though unstressed syllables generally promote the change due to reduced articulatory effort.[12][3][11] Exceptions and constraints further delimit the process. Clusters involving /l/ often resist vocalization (e.g., /lk/ in "milk" or /lp/ in "help" remain lateral to ensure distinctiveness), with preceding consonants inhibiting the change in complex codas (41.7% rate vs. 55.3% in simple codas). Diachronically, morphological factors can condition avoidance, such as in function words where paradigmatic contrasts are preserved (e.g., lower rates before morpheme boundaries). These constraints align with sonority principles, favoring vocalization only where it improves syllable well-formedness without violating faithfulness to underlying /l/.[11][1][3]Typology of L-Vocalization
Labiovelar and Back Vowel Type
The labiovelar and back vowel type of L-vocalization refers to the phonological process in which a velarized lateral approximant [ɫ] shifts to a back rounded glide or vowel such as [ʊ], , or , primarily in syllable coda positions. This transformation arises from the inherent articulatory properties of dark [ɫ], where the tongue dorsum retracts toward the velum during velarization, reducing the lateral airflow and enhancing the dorsal gesture; subsequent labialization introduces lip rounding, further aligning the segment with vocalic sonority by increasing its acoustic prominence and perceptual vowel-likeness.[13][14] Typologically, this subtype is motivated by a cross-linguistic preference for structures that maximize sonority and favor open syllables over closed ones containing obstruent-lateral sequences. By converting [ɫ] into a more sonorous back rounded element, the process resolves potential sonority violations in codas, as seen in abstract derivations like *kal > [kau], where the lateral's low sonority relative to adjacent vowels is ameliorated through vocalization. This natural progression reflects broader phonological tendencies toward articulatory ease and perceptual clarity, often observed in language acquisition and sound change.[11][1] The distribution of labiovelar L-vocalization is concentrated in Germanic languages, where dark l allophony in codas predisposes the segment to this change, and in select Romance languages exhibiting similar velarization patterns. It typically occurs preconsonantally or word-finally, with higher rates before alveolar or dental consonants due to dissimilatory effects that enhance the perceptual distinction from the lateral's coronal properties. This conditioning by dark l allophony distinguishes it from clear environments, limiting its application to languages or dialects with robust velarization.[13][1] Theoretical explanations within Optimality Theory frame this vocalization as the optimal output of interacting constraints, where markedness constraints like *LATERAL (prohibiting lateral approximants in coda or nucleus positions) and sonority-sensitive ones such as *LOW-SON-CODA (penalizing low-sonority segments in codas) dominate faithfulness constraints like IDENT-LATERAL (preserving lateral quality). A higher-ranked MAX (against deletion) ensures that vocalization to or is preferred over total segment loss, as the back rounded output better satisfies sonority hierarchies while maintaining segmental material. This ranking captures the process's variability and naturalness across languages.[11]Palatal and Front Vowel Type
The palatal and front vowel type of L-vocalization involves the realization of the clear lateral approximant as a front palatal glide or a high front vowel such as [ɪ] or , typically triggered by assimilation to adjacent front vocalic features rather than velarization. This process entails a palatalization of the coronal articulation of , leading to a loss of laterality while preserving or enhancing frontness, often in syllable-final or intervocalic positions following mid or high front vowels. Unlike velar types, it is not primarily driven by sonority enhancement but by coarticulatory assimilation, resulting in approximants that align with front vowel harmony.[15] This subtype is typologically motivated in languages lacking a phonemic dark [ɫ]- contrast, where clear predominates and vocalization reinforces front articulation to maintain perceptual harmony with surrounding vowels. It occurs preferentially after front vowels like /e/ or /i/, promoting a smooth transition via rising formant structures that perceptually cue a glide rather than a lateral constriction. The process is assimilation-based, with the [+front] feature from the vowel influencing the lateral's place node, often yielding as an intermediate stage before potential full vocalization to . In contrast to the labiovelar subtype, which involves dorsal retraction, this type emphasizes coronal frontness without place delinking to velars.[15] Distributionally, the palatal and front vowel type is widespread in Romance languages, particularly in dialects of Italian, Occitan, Catalan, and Spanish, where /l/ after front vowels vocalizes to , as in Occitan examples like *falu > [faju] or Spanish dialectal realizations post-yeísmo. Analogous patterns emerge in certain Romance varieties like Romanian influenced by front vowel contexts. These occurrences are more prevalent in systems with robust palatal contrasts, underscoring the role of front environments in driving the shift.[16] Theoretical accounts frame this vocalization within feature geometry, where the coronal node of absorbs [+front, +high] features from an adjacent vowel via spreading, followed by delinking of the [lateral] feature to yield a non-lateral approximant . This hierarchical structure under the place node allows targeted palatal assimilation without affecting other articulators, contrasting with labiovelar types that delink the coronal node for dorsal dominance. Seminal models highlight how such spreading enhances perceptual salience in front harmonic systems, explaining the subtype's prevalence in clear-l languages.Germanic Languages
Historical Developments
In Proto-Germanic, the lateral approximant /l/ remained stable as a consonant in onset positions and medially before vowels, as evidenced by consistent reflexes across daughter languages such as Gothic laggs, Old Norse langr, Old English long, and Old High German lang from PGmc. langaz 'long'. However, post-500 CE, /l/ in coda positions became prone to vocalization, particularly in sequences following back vowels, marking the onset of a diachronic process observed in emerging West Germanic varieties. A key development involved the vocalization of /l/ in sequences like -al-, -ol-, and -ul-, where it shifted to vowel or semivowel outcomes such as /u/, /au/, or /ou/, often conditioned by the preceding vowel quality. For instance, PGmc. halbaz 'half' evolved into Middle Scots hawf [hɑuf] through intermediate stages reflecting this change, while in Scots, related forms like alliz > "aw". This process aligned with the labiovelar type of L-vocalization, where coda /l/ after labial or velar contexts developed into back rounded vowels or diphthongs.[17][18] Comparatively, in Middle Dutch (12th–15th centuries), /ul/, /ol/, and /al/ merged and vocalized to /ou/ before dentals like /d/ or /t/, as seen in schulder > schouder 'shoulder' and ald > oud 'old'. Similarly, Middle Scots exhibited /al/ > /au/ around the 14th century, with examples including OE healf > hawf 'half' and OE call > caw 'call', though the change was not fully realized by 1500. These timelines highlight parallel evolutions from shared proto-forms, with vocalization progressing more rapidly in insular varieties.[19] Influencing factors included preceding vowel length, which promoted diphthongization in longer vowels (e.g., /aːl/ > /au/), and i-umlaut, which could front or alter the quality of the resulting vowel in umlauted environments. Resistance to full vocalization occurred in high-frequency words, where analogical leveling preserved /l/ longer, as observed in lexical distributions favoring conservative forms in core vocabulary.[20]Modern Varieties
In modern German, L-vocalization affects coda /l/ in colloquial standard varieties, where it is typically realized as a glide [ɪ̯] or high vowel , leading to mergers like "welche" [ˈvɛlçə] with "Seuche" [ˈzɔɪ̯çə]. This tendency is more moderate in northern dialects but contributes to informal speech patterns across urban centers. In Bavarian dialects, including those spoken in Upper Austria and Munich, post-vocalic /l/ undergoes fuller vocalization to [ɪ] or a rounded , as in "viel" [fɪə̯] or [vui̯] (much), contrasting with the retained in standard forms like [fiːl]. Such changes highlight dialectal divergence, with Bavarian varieties showing near-categorical application after back or front vowels.[8] Dutch exhibits a rising incidence of L-vocalization in standard varieties since the 1950s, primarily in non-prevocalic codas, where /l/ shifts to a dark glide [ʊ̯] or central [ə], as in "koel" [kuːɤ̯] (cool) or "melk" [mɛɫək] (milk). Usage has increased from around 20% in the 1920s to over 50% by the 1980s, with rates exceeding 80% among women born after 1981 in urban areas like Amsterdam. Rural dialects, such as those in Flanders and Limburg, display lower vocalization (under 40%), reflecting sociogeographic divides where urban centers accelerate the shift.[21] In North Germanic languages like Swedish and Norwegian, L-vocalization is rare or absent in coda positions, with /l/ typically realized as a clear or velarized [ɫ] without replacement by a vowel or glide. Sociolinguistic patterns of L-vocalization in Germanic languages reveal consistent effects of age and gender, with younger speakers—particularly females—driving adoption as a marker of innovation. In Manchester English, females produce darker /l/ variants than males, especially in lower social classes, contributing to higher vocalization rates over time. Across English dialects, women lead up to 90% of phonological changes, including this one, with 2020s data from London showing youth-led progress toward potential standardization in informal RP-influenced speech. In Dutch, post-1950s increases are spearheaded by urban women, underscoring gender as a key accelerator in urban-rural gradients.[22][23][24][21]Romance Languages
Old French and Related Changes
In Old French, L-vocalization primarily affected pre-consonantal and word-final /l/, transforming it into during the 9th to 12th centuries, a process that exemplifies the labiovelar type of this sound change. This shift began before the 10th century and became widespread by the 11th and 12th centuries, as evidenced by textual attestations in early medieval manuscripts. For instance, Latin calidus ('warm') evolved to Old French chalz and then chaut [ʃawt], ultimately yielding modern chaud [ʃo]. Similarly, dolce ('sweet') became douce [dusə].[5][25] The change occurred in intervocalic positions before a consonant or at word boundaries, but was conditioned by the preceding vowel: after non-high vowels (except /i/), /l/ vocalized to , often creating diphthongs like [au]; after /i/, however, /l/ underwent effacement rather than vocalization, as in fils [fis] from filius. Palatal environments, such as those involving /i/ or subsequent palatal consonants, typically blocked full vocalization, leading instead to deletion or palatalization. This restriction preserved /l/ in certain clusters, contributing to dialectal variations across Gallo-Romance varieties.[5][25] The phonetic outcome involved diphthongization, with [al] progressing to [au] by the late 12th century, followed by monophthongization to in subsequent centuries; comparable shifts affected other vowels, such as [ol] to [ou] > . This profoundly influenced Old French orthography, where spellings like eau emerged to represent the diphthong [au] (e.g., eau for earlier al), reflecting a lag between spoken changes and written conventions that persisted into Middle French.[5] In descendant Gallo-Romance languages, the effects of this vocalization endure in the modern French vowel system, where the resulting mid-back vowels (, ) are standard, and the original /l/ is entirely lost in affected positions. However, orthographic /l/ is sometimes retained as a silent letter, enabling its optional pronunciation in liaison contexts (e.g., tout entier [tu.tɑ̃.tje] in careful speech), though the historical /l/ itself is not revived. This legacy underscores the role of L-vocalization in simplifying French syllable structure and reducing consonant clusters.[5][25]Italo-Dalmatian and Eastern Romance
In Italo-Dalmatian languages, L-vocalization predominantly manifests as a palatal shift, where Latin /l/ in syllable codas or clusters evolves into the palatal approximant /j/, particularly in central varieties like Tuscan Italian. A representative example is the development from Latin *flōrem to Italian fiore ['fjɔ.re] 'flower', in which the /l/ within the /fl/ cluster vocalizes to /j/ as part of post-Latin phonological innovations occurring between the 12th and 14th centuries.[26][27] Similarly, Latin *clāvem yields Italian chiave ['kjɑ.ve] 'key', illustrating the process in /kl/ onsets where /l/ assimilates palatally before a following vowel.[5] This change aligns with the palatal typology of L-vocalization in Romance, favoring fronted articulations over velar ones.[28] In Eastern Romance, particularly Romanian, L-vocalization is more restricted, with /l/ generally retained as a lateral approximant, though historical losses occur in intervocalic positions, often via palatalization to /j/ and subsequent deletion rather than full vocalization to a back vowel. For instance, Latin *filius 'son' develops into Romanian fiu [fij], where /l/ is lost after palatalization. This occurs preferentially in post-front vowel environments, such as after /i/, but is less pervasive than in Western Romance varieties like Old French, where velar vocalization to dominates.[29] Romanian's relative stability of dark /l/ in other positions contributes to its conservative profile compared to Italian.[30] Regional variations within Italo-Dalmatian highlight differential retention of laterals; Sicilian dialects, for example, preserve more /l/ sounds in pre-consonantal positions than central Italian, despite evidence of early vocalization (e.g., to /u/ or /i/) in 11th-century texts like "l’autri" for 'the others'.[5] In loanwords, these patterns influence adaptation, with borrowed terms undergoing similar palatal shifts in Italian (e.g., English "million" > Italian milione with /lj/ > /j/) but retaining laterals more faithfully in Romanian contexts.[29]Iberian Romance Languages
In Iberian Romance languages, L-vocalization manifests in limited and often conditioned forms, typically involving velarization of coda /l/ to [ɫ] followed by partial realization as a labiovelar glide or vowel [u, o], contrasting with the more systematic labiovelar shifts seen in Old French. Spanish exhibits rare full L-vocalization in modern standard varieties, where /l/ generally remains a clear alveolar lateral or mildly velarized [ɫ] in coda positions across most dialects. Historical instances, however, illustrate the process, as in Latin *alter > Old Spanish autro > modern otro, where intervocalic /l/ vocalized to before further monophthongization to . In certain Latin American dialects, such as those in the Dominican Republic and Puerto Rico, word-internal coda /l/ may partially vocalize to a glide [i̯] due to reduction, though deletion or maintenance of is more common.[31] These changes are gradient and sociolinguistically variable, appearing more frequently in colloquial speech but without widespread productivity. Portuguese shows more consistent L-vocalization, especially of velarized /ɫ/ in syllable-final position, a feature that has intensified in 20th-century urban varieties. In European Portuguese, word-final /ɫ/ often reduces to in casual registers, as in mal [maw] 'bad' or sol [sow] 'sun', while standard pronunciation retains [ɫ].[32] Brazilian Portuguese extends this further, with syllable-final /l/ (including non-final codas) systematically vocalizing to or , driven by articulatory weakening of the lateral gesture; examples include mal [maw ~ mau] and Brasil [braˈziw ~ ˈzi u].[33] Acoustic and articulatory studies confirm this as a cross-dialectal innovation, more prevalent in informal urban contexts like Rio de Janeiro and São Paulo, where rates exceed 90% for coda tokens.[34] Catalan displays partial L-vocalization primarily through strong velarization of coda /l/ to [ɫ] across dialects, with further reduction to or in colloquial Balearic varieties. In Balearic Catalan (e.g., Mallorquí), words like bell 'beautiful' surface as [ˈbɛw] or [ˈbɛo], reflecting gestural overlap that weakens the lateral articulation in preconsonantal or final positions.[35] This process is conditioned by prosodic factors, such as speech rate and following back vowels, and remains variable, with formal speech favoring [ɫ]; it parallels but is less advanced than Portuguese patterns.[36]Slavic Languages
South Slavic Patterns
In South Slavic languages, L-vocalization commonly involves the transformation of coda /l/ to /o/ or , a historical sound change documented from the 14th century onward, with evidence suggesting earlier medieval developments between the 10th and 14th centuries.[37] This process is particularly prominent in Serbo-Croatian, Slovene, and Bulgarian, where it affects syllable-final positions and contributes to alternations in related forms. In Serbo-Croatian, coda /l/ systematically vocalizes to /o/, as reflected in modern spelling and pronunciation; for instance, Proto-Slavic *bělъ gradъ 'white city' yields Beograd [bêɡraːd], with *bělъ becoming beo, while the feminine counterpart bela retains /l/ due to morphological conditioning.[37] This alternation is conditioned by syllable structure, applying in masculine nominative singulars and past participles but blocked in roots or other inflected forms like genitive plurals (e.g., orla 'eagle-gen.pl.' from ora-o).[37] The change is more fully realized in Bulgarian, where coda /l/ (often realized as dark [ɫ]) vocalizes to or in most contexts, especially before back vowels or in clusters, as in малко [ˈmau̯kɔ] 'a little' from Proto-Slavic *malъko.[38] This outcome predominates across standard and dialectal varieties, with near-complete loss of consonantal in such positions by the late medieval period.[39] In Slovene, historical coda /l/ is typically pronounced as or [u̯], though spelled asWest and East Slavic Patterns
In West Slavic languages, L-vocalization primarily affects the velarized lateral approximant /ɫ/, transforming it into the labial-velar approximant , particularly in syllable codas and often following back vowels. This process is most advanced in Polish, where historical /ɫ/ has systematically vocalized to across various positions, including codas, though it originated in post-vocalic environments. For instance, the word mały ("small") is pronounced [ˈmawɨ], with the coda /ɫ/ realized as . The change began in the 16th century among lower social strata and became widespread by the 18th century, leading to a productive morphological alternation between and the clear /l/ (e.g., in verb paradigms like masculine past tense forms). This vocalization is phonetically motivated by the articulatory similarity between velarized /ɫ/ and , both involving tongue root retraction and lip rounding, and it occurs partially in consonant clusters, where /ɫ/ may retain some lateral quality before fricatives or stops. Sociolinguistically, the shift has stabilized as a standard feature of modern Polish, with no significant regional variation in urban varieties.[41] Sorbian languages exhibit variation in L-vocalization compared to Polish. In both Upper and Lower Sorbian, the grapheme ł corresponds to historical dark /ɫ/, typically realized as in standard pronunciations, primarily in codas after back vowels, though some dialects retain a velarized [ɫ] or [lˠ]. For example, mały ("small") is pronounced [ˈmawɨ] in standard Lower Sorbian. In standard Upper Sorbian, clear /l/ (often palatalized [lʲ] before front vowels) contrasts with ł , indicating complete vocalization, while dialects may realize ł as [lˠ] in some contexts. This pattern reflects Sorbian's intermediate position in West Slavic developments, where vocalization is widespread but dialectally variable, influenced by contact with German. The process shows sociolinguistic stability in contemporary Sorbian speech communities, with rural dialects more frequently preserving lateral articulation.[42][43] In East Slavic languages like Ukrainian and Belarusian, L-vocalization targets velarized /ɫ/ in specific morphological and prosodic contexts, such as codas after back vowels, resulting in or , though it is less systematic than in West Slavic varieties. Ukrainian features a clear /l/ before front vowels and dark /ɫ/ elsewhere, with vocalization occurring notably in masculine past tense endings and certain lexical items, where /ɫ/ shifts to (orthographically v, realized as a labial approximant [ʋ] or in casual speech). For example, volkъ ("wolf") evolved to vovk, pronounced [ʋɔwk], with the historical /ɫ/ becoming adjacent to the back vowel /o/. This change is partial in clusters, often preserving lateral quality before obstruents, and demonstrates sociolinguistic consistency in standard Ukrainian, though rural dialects may retain [ɫ] more robustly. Belarusian parallels this pattern, using the letter ў for (a short u-like glide) in places of historical /ɫ/, as in the past tense alternation ду́маў [ˈdumaw] ("he thought") versus ду́мала [ˈdumala] ("she thought"), and lexical forms like воўк [vɔwk] ("wolf"). Vocalization is conditioned similarly after back vowels and in codas, with limited extension to clusters, and it remains a stable feature of the standard language, minimally affected by dialectal variation.[44][45] Across West and East Slavic, these patterns contrast with the central realization of L-vocalization in southern varieties, highlighting a northern tendency toward labial approximants. The processes underscore a shared Proto-Slavic inheritance of /ɫ/, with varying degrees of completion driven by phonetic ease and morphological productivity.Other Language Families
Uralic Languages
In Uralic languages, L-vocalization refers to sound changes where the lateral approximant *l or its palatal variant *ʎ from Proto-Uralic develops into vowels or approximants, often conditioned by position and vowel harmony, a hallmark feature of the family that influences suffixation and assimilation across back and front vowels.[46] These changes occurred in proto-forms before 1000 CE, reflecting early divergences within Finno-Ugric branches during the Bronze Age expansions.[46] In Hungarian, a Ugric language, the Proto-Uralic palatal lateral *ʎ systematically vocalized to the palatal approximant /j/, a process known as yodization, as evidenced in historical forms preserved in orthography likeSemitic and African Languages
True L-vocalization (replacement of coda /l/ with a vowel or semivowel) is rare or undocumented in most Semitic and indigenous African languages, with related phenomena like epenthesis in Arabic dialects or L/R interchangeability in Bantu languages sometimes conflated in analyses. In Arabic dialects, coda /l/ is typically preserved through epenthesis to resolve syllable structure constraints, as in Hijazi Arabic where clusters involving /l/ trigger vowel insertion (e.g., /naml/ > [namil]), without /l/ becoming vocalic. This contrasts with the preservation of /l/ in onset positions across dialects.[50][51] In African languages, particularly Bantu varieties, /l/ often interchanges with /r/ due to substrate influences, but full vocalization is not systematic. Documentation remains limited, with studies noting higher rates of L-vocalization in English-influenced varieties like African American English, but fewer investigations into indigenous Bantu or Semitic-African contact zones.[52] Challenges in researching L-vocalization in Semitic and African languages include frequent conflation with L/R mergers, as seen in many Bantu languages where /l/ and /r/ alternate without distinct vocalic outcomes, complicating phonetic analysis. Additionally, incomplete coverage in non-European linguistics stems from resource gaps in sub-Saharan Africa, where sociolinguistic studies prioritize morphology over phonetics, leading to underrepresentation of dialectal variation. These issues highlight the need for targeted fieldwork to distinguish L-vocalization from related approximant shifts.References
- https://en.wiktionary.org/wiki/Reconstruction:Proto-Germanic/halbaz