Welsh orthography
Welsh orthography is the standardized system for writing the Welsh language, a Celtic language spoken primarily in Wales, using a 28-letter variant of the Latin alphabet that includes 20 single letters and eight digraphs treated as distinct units (ch, dd, ff, ng, ll, ph, rh, th).[1] This system is largely phonemic, meaning that, with few exceptions, each letter or digraph consistently represents a single phoneme, facilitating predictable pronunciation for educated speakers and contributing to the language's relative transparency in reading acquisition compared to languages like English.[2] The Welsh alphabet, known as yr wyddor, omits the letters j, k, q, v, x, and z in native words, though they appear in loanwords, and incorporates w and y as vowels alongside a, e, i, o, u.[1] Vowels can be short, long (often marked by a circumflex accent, such as â or ŷ), or medium in length, with stress typically falling on the penultimate syllable in polysyllabic words; the seven vowels (a, e, i, o, u, w, y) produce a range of sounds, including the distinctive Welsh y, which varies between [ɨ] or [ə] depending on position.[2] Consonants include unique sounds like ll (a voiceless lateral fricative, [ɬ]), rh (a voiceless trill, [r̥]), and dd (a voiced dental fricative, [ð]), with unaspirated stops (p, t, c) contrasting against their aspirated English counterparts.[1] A defining feature of Welsh orthography is its integration of initial consonant mutations—soft (e.g., p to b), nasal (e.g., t to nh), and aspirate (e.g., c to ch)—which alter word-initial sounds based on grammatical context, such as possession or number, and are reflected in spelling changes without altering the root form.[2] Additional diacritics include the grave accent for short vowels in certain positions and the diaeresis to separate adjacent vowels (e.g., aë), ensuring clarity in pronunciation.[3] These elements evolved from early medieval Latin script adaptations, with modern standardization emerging in the 16th century through works like those of William Salesbury, and continuing today via the Welsh government's Standardisation Panel, which modernizes spelling based on the authoritative Geiriadur Prifysgol Cymru dictionary to promote consistency and accessibility.[4][3]Overview
Alphabet Composition
The Welsh orthography utilizes a modified version of the Latin script, featuring a 29-letter alphabet that distinguishes it from the standard 26-letter English alphabet. This composition includes 21 single letters—A, B, C, D, E, F, G, H, I, J, L, M, N, O, P, R, S, T, U, W, Y—and eight digraphs regarded as individual letters: ch, dd, ff, ng, ll, ph, rh, th.[5] The inclusion of these digraphs as atomic units reflects the orthography's effort to represent distinct phonetic elements efficiently within the Latin framework.[6] Notably absent from native Welsh words are the letters J, K, Q, V, X, and Z, which appear primarily in loanwords or proper names borrowed from other languages; J, while part of the modern alphabet, is rarely used in original Welsh vocabulary.[5][7] Each letter and digraph exists in three forms: majuscule (uppercase, e.g., A, Ll), minuscule (lowercase, e.g., a, ll), and titlecase variants used in headings or proper nouns (e.g., Ll in "Llanelli"). For digraphs, capitalization applies only to the initial component, preserving their unity as single letters.[8][6] In practical applications, such as dictionaries and alphabetical sorting, digraphs function as indivisible units; for instance, "Llanfair" is indexed under L, following words beginning with single L but preceding those with M.[6] This treatment ensures consistency in lexicographical ordering and reinforces the digraphs' status as core alphabetic elements.[8]Core Orthographic Principles
Welsh orthography operates on a largely phonemic basis, where each phoneme is typically represented by a single letter or a consistent digraph, such as ch for /χ/, ll for /ɬ/, rh for /r̥/, and ng for /ŋ/, enabling a standardized pronunciation that is intelligible across educated speakers. This system ensures a high degree of transparency, with the spoken form predictable from the written one in most cases, though exceptions arise due to regional phonological differences. For instance, the orthography accommodates North-South dialectal variations, such as the southern distinction between short and long vowels in the penult syllable (e.g., tonau vs. tonnau), which northern speakers do not phonetically realize, as non-final vowels are short in Northern Welsh, or the southern tendency to pronounce final y as , blurring contrasts with i.[2][9] A key principle is etymological transparency, where spelling preserves historical roots even when modern pronunciation has shifted, as seen in the retention of w to denote the vowel /u/ (e.g., cwm /kuːm/, valley), a convention rooted in earlier stages of the language where u served other functions, maintaining links to Proto-Celtic origins. This approach balances phonemic consistency with historical continuity, avoiding purely sound-based reforms that could obscure etymological connections, such as the spelling aw for diphthongs that monophthongized to /oː/ in some dialects, as in historical /aʊ/ in unstressed final syllables (e.g., certain plural forms). Dialectal influences further shape this, with northern orthographic norms reflecting finer vowel gradations not always pronounced in the south.[10][2][11] The orthography plays a crucial role in distinguishing minimal pairs through targeted markings, particularly for vowel length, without disrupting the core word form. For example, tan (/tan/, until) contrasts with tân (/taːn/, fire), where the circumflex accent on the latter indicates a long vowel, preserving phonemic distinctions essential for meaning. Similarly, vowel lengths are encoded via the circumflex in monosyllabic roots (e.g., tŵr /tʊːr/, tower, vs. twr /tʊr/, heap), ensuring clarity while adhering to the root's integrity. Initial consonant mutations—soft, nasal, and aspirate—are orthographically encoded by altering the initial letter of the word (e.g., pen /pɛn/ becomes fy mhen /və mɛn/ under nasal mutation), reflecting phonological processes without modifying the underlying root spelling. This method maintains etymological stability, as the mutated form signals grammatical relationships (e.g., possession or negation) while the base form remains unchanged in dictionaries and citations. Vowel length in mutated contexts follows the same root-based rules, prioritizing consistency across dialects.[2][12]Historical Development
Early Forms and Influences
The earliest known Welsh texts date to the late 8th or early 9th century, emerging from the Brythonic languages spoken by the Britons following the Roman withdrawal from Britain. These initial writings adapted the Latin script, which had been introduced through Roman administration and Christian missionary activity, to represent Brythonic phonemes, marking the transition from oral traditions to written records. Surviving examples include marginal glosses and inscriptions, such as those in the Juvencus manuscript (9th–10th century), which demonstrate an orthographic system adapted for early Welsh phonemes.[13] Influences from Latin and Old English significantly shaped this nascent orthography. Latin provided the foundational alphabet, with adaptations for Welsh sounds, while contact with Anglo-Saxon scribes introduced runic-derived characters like thorn (⟨þ⟩) and eth (⟨ð⟩) to denote the voiceless /θ/ and voiced /ð/ dental fricatives, respectively—sounds absent in Latin but present in Brythonic. These symbols appeared in Old Welsh texts, such as the late 8th- or early 9th-century Surexit memorandum in the Lichfield Gospels, where ⟨þ⟩ renders "papeþ" for "pa beth" ("what"). Their use persisted into the late medieval period and even the 16th century, reflecting scribal borrowing from English manuscripts, though Welsh writers often interchanged them or reverted to simple ⟨d⟩ for /ð/.[3][14] Medieval developments saw gradual refinements, including the introduction of digraphs to distinguish sounds more clearly. By the 13th century, ⟨dd⟩ began appearing sporadically for /ð/, as in late 12th-century examples like "dy ẟiweẟ" ("thy end"), but it gained prominence in printed works. The 1567 Welsh New Testament, translated primarily by William Salesbury, marked a key milestone by systematically employing ⟨dd⟩—replacing earlier symbols like ⟨ð⟩—to represent this voiced fricative, as seen in words like "newydd" ("new"). This innovation, influenced by Salesbury's scholarly aim to align Welsh with classical languages while preserving native phonology, helped bridge medieval inconsistencies toward greater uniformity.[3][15] Pre-19th-century Welsh spelling exhibited considerable variability, particularly in manuscripts of poetry and prose, due to regional dialects, scribal preferences, and the lack of standardization. Vowel representations were especially inconsistent; for instance, the high front vowel /ɨ/ and schwa /ə/ were often both spelled ⟨y⟩, leading to ambiguities in texts like the Black Book of Carmarthen (c. 1250), where forms such as "mynwgyl" (for "mwnwgl") show epenthetic vowels inserted for metrical purposes. Consonant spellings also fluctuated, with ⟨u, v, w⟩ interchangeably for /v/ (e.g., "niuer" for "nifer" in the Book of Aneirin, c. 13th century) and ⟨k⟩ or ⟨c⟩ for /k/. Poetry manuscripts, such as those containing the Gododdin, further illustrate this through irregular vowel shifts, like ⟨y⟩ for ⟨i⟩ in "ffuryf" (for "ffurff"), underscoring the orthography's evolution from a flexible, phonetically approximate system to one seeking precision amid diverse influences.[3][13]Modern Standardization
In the 19th century, efforts to unify Welsh spelling intensified amid growing literacy and print culture, with scholars promoting consistent use of digraphs to reflect phonetic values more reliably. Daniel Silvan Evans's An English and Welsh Dictionary (1852–1858) contributed to this by standardizing spellings in a major reference work.[16] These reforms built on earlier efforts to align orthography with pronunciation. A pivotal advancement came with John Morris-Jones's lifelong campaign to standardize Welsh orthography, beginning with articles in Y Beirniad (1890) and Y Geninen (1891–1892), followed by pamphlets like Welsh Orthography (1896) and A Guide to Welsh Orthography (1905).[17] His 1928 book Orgraff yr Iaith Gymraeg synthesized these ideas into a comprehensive framework, drawing on historical linguistics to establish consistent rules for letters, digraphs, and mutations, which became the basis for modern Welsh spelling.[17] This work profoundly impacted education and literature: it was integrated into school curricula across Wales, shaping how Welsh was taught in primary and secondary schools, and adopted by publishers as the authoritative standard for books, newspapers, and official documents, fostering a unified literary language.[18] Further refinements occurred in 1987, when a committee chaired by Stephen J. Williams recommended minor updates to accommodate loanwords, officially introducing the letter ⟨j⟩ to represent the sound /dʒ/ (as in English "jam"), thereby expanding the traditional 28-letter alphabet to 29 while maintaining phonetic consistency. Post-1987 developments have included ongoing discussions about accommodating dialectal variations in spelling, particularly for northern and southern pronunciations of vowels like /ɨ/ and /ə/, though the Morris-Jones standard remains dominant in formal writing.[19] Additionally, advancements in digital encoding have ensured robust support for Welsh characters in Unicode since the early 1990s, with updates through the 2020s improving collation and display for accented letters and digraphs in computing and web applications. In recent years, the Welsh Government's Standardisation Panel has continued this work, modernizing spelling and promoting consistency based on the Geiriadur Prifysgol Cymru, with consultations as recent as 2021.[4]Letters and Phonetic Values
Single Letters
The Welsh alphabet consists of 20 single letters: A, B, C, D, E, F, G, H, I, L, M, N, O, P, R, S, T, U, W, and Y. These letters represent distinct phonetic values, primarily monophthongs for the vowels (A, E, I, O, U, W, Y) and consonants for the others, with variations between northern and southern dialects most pronounced in the high central vowels and the realization of R. The letter names follow traditional Welsh conventions, such as "a" for A and "bi" for B. Pronunciations are largely consistent across dialects for consonants, but vowels like U and Y differ notably, with northern Welsh featuring the close central [ɨ] where southern uses near-close front [ɪ]. The following table details each single letter, including its name, primary IPA phonetic values (distinguishing northern and southern where applicable), English approximations, and representative example words with approximate pronunciations. Vowel lengths are indicated where relevant, as length affects quality but is determined by orthographic rules elsewhere. Letters like J, K, Q, V, X, Z appear in loanwords but are not part of the core alphabet.| Letter | Name | Northern IPA | Southern IPA | English Approximation | Example Word (Northern/Southern) |
|---|---|---|---|---|---|
| A | a | /aː/ (long), /a/ (short) | /aː/ (long), /a/ (short) | father (long), cat (short) | â /aː/ in bâch [baːχ] "small"[]; a /a/ in can [kan] "song"[] |
| B | bi | /b/ | /b/ | bat | bac [bak] "stick"[] |
| C | ci | /k/ | /k/ | cat | car [kar] "love"[] |
| D | di | /d/ | /d/ | dog | dŷ [dɨː] "house"[] |
| E | e | /ɛː/ (long), /ɛ/ (short) | /ɛː/ (long), /ɛ/ (short) | air (long), bet (short) | ê /ɛː/ in mel [mɛːl] "mule"[]; e /ɛ/ in pen [pɛn] "head"[] |
| F | ef | /v/ | /v/ | van | ef [ɛv] "he"[] |
| G | gi | /ɡ/ | /ɡ/ | go | gân [ɡaːn] "song"[] |
| H | aitch | /h/ | /h/ | hat | heno [hɛnɔ] "tonight"; used mainly in aspirated mutations or loanwords[] |
| I | i | /iː/ (long), /ɪ/ (short) | /iː/ (long), /ɪ/ (short) | see (long), sit (short) | î /iː/ in mîn [miːn] "fine"[]; i /ɪ/ in mil [mɪl] "thousand"[] |
| L | el | /l/ | /l/ | love | llyfr [ɬɪvr] "book"[] |
| M | em | /m/ | /m/ | man | mam [mam] "mother"[] |
| N | en | /n/ | /n/ | no | nos [nɔs] "night"[] |
| O | o | /ɔː/ (long), /ɔ/ (short) | /ɔː/ (long), /ɔ/ (short) | law (long), hot (short) | ô /ɔː/ in môr [mɔːr] "sea"[]; o /ɔ/ in pot [pɔt] "pot"[] |
| P | pi | /p/ | /p/ | pen | papyr [papɪr] "paper"[] |
| R | er | /r/ (trilled) | /ɾ/ (tapped) | red (rolled in north, tapped in south) | bara [ˈbaɾa] / [ˈbaɾa] "bread"[]; dialectal variation affects trill strength[] |
| S | es | /s/ | /s/ (or /ʃ/ before i/y) | sun (or shin before i/y in south) | sêr [seːr] "star"; southern lisping tendency before i/y[] |
| T | ti | /t/ | /t/ | top | tŷ [tɨː] / [tiː] "house"[] |
| U | u | /ɨː/ (long), /ɨ/ (short) | /iː/ (long), /ɪ/ (short) | roses (north long), see (south long) | u /ɨː/ in un [ɨːn] "one" (north) / [iːn] (south)[]; u /ɨ/ in pum [pɨm] "five" (north) / [pɪm] (south)[] |
| W | wi | /uː/ (long), /ʊ/ (short) | /uː/ (long), /ʊ/ (short) | boot (long), book (short) | ŵ /uː/ in mŵg [muːɡ] "mug"[]; w /ʊ/ in gwell [ɡwɛɬ] "better"[] |
| Y | y | /ə/ (unstressed), /ɨː/ (long stressed), /ɨ/ (short stressed) | /ə/ (unstressed), /iː/ (long stressed), /ɪ/ (short stressed) | the (unstressed), bit (stressed north short), sit (south short) | y /ə/ in ysgol [əsɡɔl] "school"[]; ŷ /ɨː/ in dŷ [dɨː] "house" (north) / [diː] (south)[]; y /ɨ/ in dyn [dɨn] "man" (north) / [dɪn] (south)[] |
Digraphs
In Welsh orthography, eight consonant digraphs are treated as distinct single letters within the 28-letter alphabet, each representing a unique phoneme not conveyed by single letters alone.[20] These digraphs—ch, dd, ff, ng, ll, ph, rh, and th—originate from historical developments in the language and are essential for accurately rendering Welsh phonology.[21] Unlike English digraphs, which often combine to form new sounds from adjacent letters, Welsh digraphs function atomically, with their own names, positions in the alphabet, and orthographic behaviors.[3] The following table summarizes the digraphs, their traditional names (as used in early 20th-century grammars), International Phonetic Alphabet (IPA) values, English approximations where applicable, and representative examples:| Digraph | Name | IPA | Approximation | Example |
|---|---|---|---|---|
| ch | ech | /χ/ | Scottish "loch" or German "Bach" | chwech (/χwɛχ/, "six")[21][3][1] |
| dd | edd | /ð/ | Voiced "th" in "this" | ddau (/ðaɪ/, "two")[21][3][1] |
| ff | eff | /f/ | "f" in "off" | ffwrdd (/fʊrð/, "away")[21][3][1] |
| ng | eng | /ŋ/ | "ng" in "sing" | bangor (/ˈbaŋɡɔr/, "choir")[21][3][1] |
| ll | ell | /ɬ/ | Voiceless "l" (hissed with tongue against teeth) | llan (/ɬan/, "enclosure")[21][3][1] |
| ph | fi | /f/ | "f" in "phone" | phrif (/friːv/, "chief")[21][3][1] |
| rh | rhī | /r̥/ | Voiceless trill (like "hr" with breath) | rhy (/r̥ɨ/, "very")[21][3][1] |
| th | eth | /θ/ | Unvoiced "th" in "think" | tair (/θair/, "three")[21][3][1] |
Diphthongs
Welsh orthography consistently represents diphthongs through sequences of two vowel letters, capturing both rising and falling articulations without dedicated diacritics or special characters beyond the standard alphabet. This system aligns closely with phonetic realizations, allowing for straightforward reading once dialectal variations are accounted for.[22] The Welsh language features twelve principal diphthongs, whose pronunciations exhibit notable differences between Northern and Southern varieties, particularly in the quality of the initial vowel element. In Northern Welsh, diphthongs often begin with a more open or central vowel, while Southern forms tend toward simplification or monophthongization in casual speech. The following table summarizes these diphthongs, including their standard orthographic forms, International Phonetic Alphabet (IPA) transcriptions with dialectal notes, English approximations, and illustrative examples.[23][22]| Orthography | IPA (Northern/Southern) | English Approximation | Example (Word/Meaning) |
|---|---|---|---|
| ae | /aɪ/ | eye | caeth (/kaɪθ/, captive) |
| ai | /aɪ/ | eye | tair (/taɪr/, three) |
| au | /aʊ/ | out (Northern); /aʊ/ or /aː/ (Southern) | haul (/haʊl/, sun) |
| aw | /aʊ/ | out | baw (/baʊ/, ball) |
| ei | /əɪ/ or /eɪ/ (Northern); /eɪ/ (Southern) | eight (Northern) or day (Southern) | beic (/bəɪk/ or /beɪk/, bike) |
| eu | /ɛʊ/ or /eɪ/ | (no direct; like "air" gliding to "oo") | heul (/heɪl/, sun; homophone with haul in some dialects) |
| ew | /ɛʊ/ or /ju/ | (like "e" in "bed" to "oo"); or "you" | ewin (/ɛwɪn/, anvil) |
| oe | /ɔɪ/ | boy | moel (/mɔɪl/, bald) |
| oi | /ɔɪ/ | boy | toiled (/tɔɪlɛd/, pierced) |
| ou | /ɔʊ/ | (like "aw" in "law" to "oo") | bou (/bɔʊ/, cow; dialectal) |
| uw | /ɪʊ/ | (like "ee" to "oo") | duw (/dɪʊ/, God) |
| wy | /ʊɪ/ | (like "oo" to "ee"; rounded lips) | gwynt (/ɡʊɪnt/, wind) |
Special Orthographic Rules
Diacritics and Accents
In Welsh orthography, the circumflex (ˆ), referred to as to bach ("little roof"), is the most common diacritic and serves to indicate long vowels in positions where they would otherwise be interpreted as short, thereby distinguishing lexical items and ensuring accurate pronunciation. For instance, tân (/ta:n/, "fire") contrasts with tan (/tan/, "under"), with the circumflex marking the elongated vowel in the former. This usage is mandatory in formal writing, such as official documents and literature, to resolve ambiguities arising from orthographic context, and it also signals stress on the marked syllable.[8][24] The acute (´) and grave (`) accents are far less frequent, appearing primarily in dictionaries and phonological analyses to denote stress patterns or vowel quality deviations from standard expectations. The acute accent highlights stress on a syllable outside the default penultimate position, as in casáu (/kasa.u/, "to hate"), where it overrides the typical stress rule. The grave accent, meanwhile, may mark short vowels in environments predicting length, such as siòl (/ʃɔl/, "skull") or dictionary notations like è for /ɛ/. These marks are lexical exceptions rather than routine features of prose, reflecting irregular prosody in borrowed or specialized terms.[24] The diaeresis (¨) functions to separate contiguous vowels, preventing their coalescence into a diphthong and clarifying syllabification, particularly in compound words or place-names. It is commonly applied to i between vowels, as in copïo (/kɔ.pi.jɔ/, "to copy") or Cwmsyfïog (a place-name requiring separation for the penultimate i). Usage is obligatory when the following element exceeds two syllables to avoid mispronunciation, but optional in shorter ambiguous cases like Gïas. This diacritic supports the orthography's phonetic transparency without altering core vowel length rules.[8] Diacritics are generally omitted when vowel length or stress can be reliably inferred from positional conventions, such as final syllables inherently bearing length, limiting their application to exceptional cases for precision in formal contexts.[24]Vowel Length Determination
In Welsh orthography, vowel length is primarily determined by the position of the stressed vowel relative to surrounding consonants, with long vowels occurring in specific contexts that follow predictable phonological patterns.[25] Stressed vowels are typically long when they appear in an open syllable at the end of a word or before a single voiced stop consonant such as /b/, /d/, or /g/, as in beg pronounced /bɛːɡ/ ("claim"), where the vowel lengthens before the voiced /ɡ/.[24] In contrast, vowels remain short before voiceless stops like /p/, /t/, or /k/, or in similar positions without the lengthening trigger, as seen in pen /pɛn/ ("head"), where the vowel stays short before the following nasal.[25] Exceptions to this primary rule arise with geminate consonants (such as nn or rr) or certain consonant clusters, where stressed vowels are consistently short regardless of the consonant's voicing; for instance, onn /ɔn/ ("ash trees") features a short vowel before the geminate /nː/.[24] Additionally, vowels before sonorants like /l/, /m/, /n/, or /r/ can vary in length and must often be learned lexically, though they tend toward shortness in non-final positions.[25] Dialectal variations influence these patterns, with northern Welsh dialects exhibiting more pronounced length contrasts, particularly restricting long vowels to final stressed syllables, while southern varieties allow lengthening in penultimate syllables under similar conditions.[26] Vowel length is generally unmarked in standard orthography unless the context creates ambiguity, relying on an algorithmic prediction based on syllable structure and consonant type to infer phonology from spelling.[24] This involves context-sensitive rules that scan for lengthening environments, such as single voiced stops or word-final position, applying length by default in those cases while defaulting to shortness before clusters or geminates.[24] When the predicted length deviates from the standard—such as requiring a long vowel in a typically short context—a circumflex accent (^) overrides the rule to indicate length explicitly, as in môr /moːr/ ("sea"), where the mark ensures the long pronunciation despite potential ambiguity.[25] For example, march /marχ/ ("horse") follows the default short vowel before the fricative /χ/, but diacritics like the circumflex can be used sparingly for non-standard lengths in ambiguous cases.[24]Initial Consonant Mutations
Initial consonant mutations represent a core grammatical mechanism in Welsh, altering the initial consonant of words in response to preceding triggers such as possessives, prepositions, and certain syntactic structures. These changes are systematically encoded in the orthography through the substitution of specific letters or digraphs, reflecting the phonetic shifts while adhering to standardized spelling rules established in modern Welsh. Unlike some languages that use diacritics or abbreviations for such alternations, Welsh spells out the full mutated forms, ensuring that mutations are visually distinct and integral to readability.[27][28] The three primary types of initial consonant mutations are soft mutation, nasal mutation, and aspirate mutation, each triggered by distinct grammatical contexts and affecting a subset of the radical (unmutated) consonants. Soft mutation, the most frequent, involves lenition or weakening of voiceless stops to voiced stops, fricatives to approximants, and nasals or lateral fricatives to their continuant counterparts. For example, the word pen ("head") undergoes soft mutation to ben after certain prepositions like i ("to"), as in i Ben (to Head, a name). Nasal mutation, rarer and limited to specific possessive and prepositional triggers, nasalizes stops and fricatives, such as ty ("house") becoming nhy after fy ("my"), yielding fy nhy. Aspirate mutation adds aspiration to voiceless stops or introduces h-prothesis before initial vowels, for instance, car ("car") mutating to char after ei ("her"), as in ei char. These orthographic shifts draw on the standard letters and digraphs of Welsh, such as mh, nh, ngh, ph, th, and ch.[27][29][28] Mutations are triggered primarily by grammatical elements like clitic possessives (fy, ei), prepositions (ar, yn, i), numerals (dau, tri), and syntactic positions such as direct objects of finite verbs or predicate adjectives. Soft mutation occurs after approximately 47 lexical triggers, including feminine nouns with the definite article y and adverbial phrases, while nasal mutation is confined to fy and the preposition yn ("in"), and aspirate mutation follows ei (possessive "her/its") or the conjunction a in certain contexts. Syntactic triggers include adjacency to finite verbs for accusative marking in noun phrases. The following tables summarize the orthographic correspondences for each mutation type, based on the radical forms of consonants:[27][28][29] Soft Mutation| Radical | Mutated | Example (Radical → Mutated) |
|---|---|---|
| p | b | pen → ben ("head") |
| t | d | tad → dad ("father") |
| c | g | car → gar ("car") |
| b | f | bach → fach ("small") |
| d | dd | drws → ddrws ("door") |
| g | Ø (zero) | gŵr → ŵr ("man") |
| m | f | mam → fam ("mother") |
| ll | l | llan → lan ("church") |
| rh | r | rhe → re ("rhyme") |
| Radical | Mutated | Example (Radical → Mutated) |
|---|---|---|
| p | mh | pen → mhen ("head") |
| t | nh | ty → nhy ("house") |
| c | ngh | car → nghar ("car") |
| b | m | bach → mach ("small") |
| d | n | drws → nrws ("door") |
| g | ng | gŵr → ngŵr ("man") |
| Radical | Mutated | Example (Radical → Mutated) |
|---|---|---|
| p | ph | pen → phen ("head") |
| t | th | ty → thy ("house") |
| c | ch | car → char ("car") |
| Ø (vowel) | h-prothesis | ar → har ("on") |