Editing Voiced palatal approximant (section)

==Phonetic ambiguity and transcription usage==
Some languages, however, have a palatal approximant that is unspecified for rounding and so cannot be considered the semivocalic equivalent of either {{IPA|[i]}} or its rounded counterpart, {{IPAblink|y}}, which would normally correspond to {{IPAblink|ɥ}}. An example is [[Spanish language|Spanish]], which distinguishes two palatal approximants: an approximant semivowel {{IPA|[j]}}, which is always unrounded (and is a phonological vowel - an allophone of {{IPA|/i/}}), and an approximant consonant unspecified for rounding, {{IPA|[ʝ̞]}} (which is a phonological consonant). Eugenio Martínez Celdrán describes the difference between them as follows (with audio examples added):{{sfnp|Martínez Celdrán|2004|p=208}}

{{blockquote|{{IPA|[j]}} is shorter and is usually a merely transitory sound. It can only exist together with a full vowel and does not appear in syllable onset. [On the other hand,] {{IPA|[ʝ̞]}} has a lower amplitude, mainly in F2. It can only appear in syllable onset. It is not noisy either articulatorily or perceptually. {{IPA|[ʝ̞]}} can vary towards {{IPAblink|ʝ}} in emphatic pronunciations, having noise (turbulent airstream). (...)

There is a further argument through which we can establish a clear difference between {{IPA|[j]}} and {{IPA|[ʝ̞]}}: the first sound cannot be rounded, not even through co-articulation, whereas the second one is rounded before back vowels or the back semi-vowel. Thus, in words like ''viuda'' {{Audio-IPA|Es-viuda.ogg|[ˈbjuða]|help=no}} 'widow', ''Dios'' {{Audio-IPA|Es-Dios.ogg|[ˈdjos]|help=no}} 'God', ''vio'' {{Audio-IPA|Es-vio.ogg|[ˈbjo]|help=no}} 's/he saw', etc., the semi-vowel {{IPA|[j]}} is unrounded; if it were rounded, a sound that does not exist in Spanish, {{IPAblink|ɥ}}, would appear. On the other hand, {{IPA|[ʝ̞]}} is unspecified as far as rounding is concerned and it is assimilated to the labial vowel context: rounded with rounded vowels, e.g. ''ayuda'' {{Audio-IPA|Es-ayuda.ogg|[aˈʝ̞ʷuð̞a]|help=no}} 'help', ''coyote'' {{Audio-IPA|Es-coyote.ogg|[koˈʝ̞ʷote]|help=no}} 'coyote', ''hoyuelo'' {{Audio-IPA|Es-hoyuelo.ogg|[oˈʝ̞ʷwelo]|help=no}} 'dimple', etc., and unrounded with unrounded vowels: ''payaso'' {{Audio-IPA|Es-payaso.ogg|[paˈʝ̞aso]|help=no}} 'clown', ''ayer'' {{Audio-IPA|Es-ayer.ogg|[aˈʝ̞eɾ]|help=no}} 'yesterday'.|author=|title=|source=}}

He also considers that "the IPA shows a lack of precision in the treatment it gives to approximants, if we take into account our understanding of the phonetics of Spanish. {{IPA|[ʝ̞]}} and {{IPA|[j]}} are two different segments, but they have to be labelled as voiced palatal approximant consonants. I think that the former is a real consonant, whereas the latter is a [[semi-consonant]], as it has traditionally been called in Spanish, or a semi-vowel, if preferred. The IPA, though, classifies it as a consonant."{{sfnp|Martínez Celdrán|2004|p=206}}

There is a parallel problem with transcribing the [[voiced velar approximant]].

The symbol {{angbr IPA|ʝ̞}} may not display properly in all browsers. In that case, {{angbr IPA|ʝ˕}} should be substituted.

In the writing systems used for most languages in Central, Northern, and Eastern Europe, the letter ''j'' denotes the palatal approximant, as in [[German language|German]] {{Langx|de|Jahr|label=none}} 'year', which is followed by IPA. Although it may be seen as counterintuitive for English-speakers, there are a few words with that orthographical spelling in certain loanwords in English like Hebrew "[[hallelujah]]" and German "[[Jägermeister]]".

In grammars of [[Ancient Greek]], the palatal approximant, which was lost early in the [[history of Greek]], is sometimes written as {{angbr|ι̯}}, an [[iota]] with the [[inverted breve]] below, which is the nonsyllabic diacritic or marker of a [[semivowel]].{{sfnp|Smyth|1920|p=11}}

There is also the '''post-palatal approximant'''<ref>Instead of "post-palatal", it can be called "retracted palatal", "backed palatal", "palato-velar", "pre-velar", "advanced velar", "fronted velar" or "front-velar". For simplicity, this article uses only the term "post-palatal".</ref> in some languages, which is articulated slightly more back than the place of articulation of the prototypical palatal approximant but less far back than the prototypical [[velar approximant]]. It can be considered the semivocalic equivalent of the [[close central unrounded vowel]] {{IPA|[ɨ]}}The International Phonetic Alphabet does not have a separate symbol for that sound, but it can be transcribed as {{angbr IPA|j̠}}, {{angbr IPA|j˗}} (both symbols denote a [[Relative articulation#Advanced and retracted|retracted]] {{angbr IPA|j}}), {{angbr IPA|ɰ̟}} or {{angbr IPA|ɰ˖}} (both symbols denote an [[Relative articulation#Advanced and retracted|advanced]] {{angbr IPA|ɰ}}). The equivalent X-SAMPA symbols are <code>j_-</code> and <code>M\_+</code>, respectively. Other possible transcriptions include a centralized {{angbr IPA|j}} ({{angbr IPA|j̈}} in the IPA, <code>j_"</code> in X-SAMPA), a centralized {{angbr IPA|ɰ}} ({{angbr IPA|ɰ̈}} in the IPA, <code>M\_"</code> in X-SAMPA) and a non-syllabic {{angbr IPA|ɨ}} ({{angbr IPA|ɨ̯}} in the IPA, <code>1_^</code> in X-SAMPA).

For the reasons mentioned above and in the article [[velar approximant]], none of those symbols are appropriate for languages such as Spanish, whose post-palatal approximant ''consonant'' (not a ''semivowel'') appears as an allophone of {{IPA|/ɡ/}} before [[front vowel]]s and is best transcribed {{angbr IPA|ʝ̞˗}}, {{angbr IPA|ʝ˕˗}} (both symbols denote a [[Relative articulation#Raised and lowered consonants|lowered]] and retracted {{angbr IPA|ʝ}}), {{angbr IPA|ɣ̞˖}} or {{angbr IPA|ɣ˕˖}} (both symbols denote a lowered and advanced {{angbr IPA|ɣ}}). The equivalent X-SAMPA symbols are <code>j\_o_-</code> and <code>G_o_+</code>.

Especially in [[broad transcription]], the post-palatal approximant may be transcribed as a palatalized velar approximant ({{angbr IPA|ɰʲ}}, {{angbr IPA|ɣ̞ʲ}} or {{angbr IPA|ɣ˕ʲ}} in the IPA, <code>M\'</code>, <code>M\_j</code>, <code>G'_o</code> or <code>G_o_j</code> in X-SAMPA).

A voiced alveolar-palatal approximant is attested as phonemic in the Huastec language.