Template:ang-IPA/documentation

	Documentation for Template:ang-IPA.
	This page contains usage information, categories, interwiki links and other content describing the template.

This template automatically generates pronunciation for Old English words, relying on Module:ang-pron.

NOTE: This documentation is incomplete. Please contact User:Benwing2 for questions.

Generally, you should supply the regular spelling of the word in |1=, properly marked up in lower case, with macrons for long vowels, and ċ ġ ċġ sċ for soft versions of these consonants. The module will automatically attempt to generate stress in the correct places, taking into account known prefixes and the |pos= parameter (see below). In addition, the following symbols can be used:

• acute accent on a vowel to override the position of primary stress (e.g. á ī́ ḗa)
• grave accent to add secondary stress (e.g. à ī̀ ḕa)
• circumflex to force no stress on the word or prefix, for example in a compound (e.g. â ī̂ ē̂a)
• . (period) to force a syllable boundary
• - (hyphen) to force a prefix/word or word/word boundary in a compound word; the result will be displayed as a single word but the consonants on either side treated as if they occurred at the beginning/end of the word
• + (plus) is the opposite of -; it forces a prefix/word or word/word boundary to NOT occur when it otherwise would
• < (less-than) to indicate a boundary between a prefix and the main stem; this is similar to - but the prefix will always be unstressed
• > (greater-than) to indicate a boundary between the main stem and a suffix; this is similar to - but the suffix will always be unstressed
• (a single char between brackets) to force a particular allophone of /θ s f ɡ x i/ (see below)

For optimal results, you may need to specify the part of speech using |pos=, either |pos=noun, |pos=adj (equivalent to |pos=noun), |pos=verb, or |pos=verbal (for participles, verbal nouns in -ing/-ung, and words formed from participles or verbal nouns by adding -nes or variants, -līċ or -līċe). The part of speech determines the recognized prefixes (see below) and the handling of those prefixes. Specifically, prefixes attached to verbs normally have no stress or secondary stress, while prefixes attached to nouns normally have primary stress, with the exception of ġe-, be- and for-. Words with part of speech verbal behave like verbs except with respect to the prefix un-, which bears primary stress in "verbals" but is unstressed in verbs.

|1=, |2=, etc.

One or more respelled words or phrases, for which pronunciation will be generated.

|pos=

Part of speech (see above). This controls the handling of prefixes. Normally, this does not need to be specified, as it is autodetected.

|ann=

If specified, add a boldface "annotation" before the generated pronunciation(s). This should be used when there are multiple pronunciations listed in an etymology section that differ in macrons and/or presence or absence of ċ or ġ. See gripe/grīpe for an example. The annotation identifies the spelling of the particular pronunciation(s) listed to the right. If |ann=1 is given, the annotation is generated from the respelling, with special respelling characters and stress accents removed. If any other value is given, it is used directly.

The module will automatically recognize and split off common prefixes and suffixes. "Split off" means that a strong morpheme boundary will be inserted between the prefix/suffix and main component. This causes /f/, /g/, /s/, /þ/, /x/ next to the boundary to be treated as if they are at the beginning or end of a word, and ensures that the main component gets its own stress independently of the prefixes and suffixes (secondary stress a prefix has primary stress, otherwise primary stress). Multiple stacked prefixes/suffixes will be correctly split off if they occur.

The prefixes and suffixes that are recognized, and their stress behavior, depend on the inferred or explicitly specified part of speech of the word. There are three basic parts of speech used for this purpose: noun (which includes adjectives and other parts of speech besides verbs), verb, and verbal. "Verbals" include participles in -ed, -od or -en and verbal nouns in -ing or -ung, as well as words formed from participles or verbal nouns by adding -nes or variants, -līċ or -līċe. Examples of the latter are ābroþennes, āberendlīċ, and ġetācniġendlīce. "Verbals" are treated as verbs for the purpose of prefix separation and stress assignment except in the case of un-, where nouns and verbals get primary stress on the prefix while verbs get secondary stress on the prefix.

Normally you do not need to specify the part of speech explicitly. It is inferred based on the following rules:

• Words ending in -an, -ān, -ōn, -ēon, and -enne are inferred to be verbs.
• Words ending in -end, -ende, -ed, -od, -en, -ing, -ung, or words ending one of these suffixes followed by one or more occurrences of any of the suffixes -l, -le, -ns or -nss (e.g. ācennedlīċ (“native”), tōmearcodnes (“description”) and unġedafenlīcnes (“unfitness, inconvenience”)) are inferred to be verbals.
• All other words are inferred to be nouns.

Prefixes are not recognized in some circumstances, specifically:

• Some prefixes are only recognized for certain parts of speech (see table below).
• Some prefixes have a restriction that they can't be directly followed by a, o, ā or ō (this applies in particular to prefixes ending in -e, such as ġe-, and prevents e.g. ġearwian or beofor from being recognized as containing a prefix).
• Prefixes with any special characters or accents in them aren't recognized.
• Prefixes aren't recognized if directly followed by + (which indicates a non-boundary).
• Prefixes aren't recognized if not followed by at least one vowel (hence betst is not recognized as containing be- and ford is not recognized as containing for-).
• Prefixes aren't recognized if not followed by at least three letters (hence inne is not recognized as containing in-).
• Prefixes aren't recognized if directly followed by a cluster of consonants that are not an allowed word onset (hence āxian is not recognized as containing ā-, because x- is not a valid onset; likewise bennian and bercan are not recognized as containing be-).
• Prefixes aren't recognized if directly followed by one or more consonants plus an explicit syllable division marker . (since splitting off the prefix would result in a syllable with no vowels).

Suffixes aren't recognized in similar circumstances, specifically:

• Some suffixes are only recognized for certain parts of speech (see table below).
• Suffixes with any special characters or accents in them aren't recognized.
• Suffixes aren't recognized if directly preceded by + (which indicates a non-boundary).
• Suffixes aren't recognized if not preceded by at least one vowel.
• Suffixes aren't recognized if the preceding text ends in an explicit syllable division marker . followed by at least one consonant (since splitting off the suffix would result in a syllable with no vowels).

The following is the complete list of prefixes recognized as of Dec 15, 2019:

Note the general pattern where prefixes are stressed as nouns but unstressed as verbs (where multisyllabic prefixes generally get secondary stress), and the fact that be-, for-, and ġe- are exceptions in that they never receive stress. When stress is assigned to a multisyllabic prefix, it goes on the first syllable.

The following is the complete list of suffixes recognized as of Dec 15, 2019:

Occasionally you will need to force a particular allophone of a multi-allophone phoneme such as /s/ to appear. An example is verbs ending in -sian. These are special-cased to have the allophone even following a voiced sound, which applies in most cases, but clǣnsian (“to cleanse”) and derivatives are exceptions, and have . To force this, use a respelling with , e.g. clǣnian.

Another example is verbs ending in -rian. If directly preceded by a short front vowel, it is assumed to be weak class I and have phonemic /j/ for written i. However, some verbs of this form are weak class II (e.g. beterian), and are pronounced with phonemic /i/. To force this, use a respelling with , e.g. be+teran (we also need the + symbol to prevent be- from being interpreted as a prefix). The following explicit-allophone symbols are currently recognized:

• (voiceless allophone of /θ/)
• (voiced allophone of /θ/)
• (voiceless allophone of /s/)
• (voiced allophone of /s/)
• (voiceless allophone of /f/)
• (voiced allophone of /f/)
• (stop allophone of /ɡ/)
• (fricative allophone of /ɡ/)
• (velar fricative allophone of /x/)
• (palatal fricative allophone of /x/)
• (glottal fricative allophone of /x/)
• (interpretation of written i as /i/ not /j/)

There is no ; use ġ to force a /j/.