PHY104 - Distinctive Features II

This second E-Lecture on Distinctive Features discusses the most widely used features of our time. It examines the main vocalic and consonantal features and shows how phonological rules and phonotactic statements can most efficiently be made using distinctive features and defining natural classes.

[@oer-vlc]

Mathematically, you are right.If features have "no meaning", i.e.they +/- X, Yand Z, everything would be fine. However, phonologists (e.g. Roca/Johnson) prefer "semantic" features, i.e. features that are meaningful. In their interpretation, Cardinals 1, 8, 9, and 16 are the only +HIGH vowels and Cardinaly 4,5,12 and 13 are the only + LOW vowels. So they don't use all 8 +HIGH values but 4 +HIGH and 12 - HIGH values (similarly with +/- LOW). That's why an additional "semantic feature is required.

[@abdullahkng7057]

Can I explain in Arabic so that I can understand that you know the conditions in the country

[@oer-vlc]

If you have just 4 vowels you are absolutely right in having just 2 features, and it would not matter what you call them. So the feature vowel height with 4 values alone can easily be encoded with 2 features. But we have 16 Cardinal Vowels. Hence we need 5 features in total.

[@oer-vlc]

As usual we will take up all questions at the end of this month in our "Questions of the month Video". JH

[@eminememinemful]

You're Such An Amazing Professor. Your Vids Are Really Of Great Help. Thank You.

[@doubtrecordsinternational]

Great video! Thank you so much. One question though: shouldn't the features in the 'Coalescent Assimilation in RP' example change to +distributed in addition to the change to -anterior? Cheers

[@melfisa]

the feature for vowel height seems really weird to programmer like me. whereas in your example «high» and «low» seem to be redundant (they always have opposite values +h-l vs -h+l) four levels can be easily encoded with two bits of information «high» and «medium» +h-m +h+m -h+m -h-m, or even «high» and «"atr"» +h+a +h-a -h+a -h-a

[@melfisa]

thanks for these lectures and additional explanations. I find them very illuminating and interesting

[@oakleywyatt1717]

Where does the equation come from? Why do we need to calculate the amount of distinctive features we need to measure sounds and why is it determined by this particular equation?

[@BeranjakGede]

I always wondering what does 'the phonological redundancies' means ? can you give an example ?

[@m.chakirly5276]

I didn't get it from 11:12 further about 'anterior'. if labials are included as anteriors, why is it that /p b/ sound are not marked with + ?

[@bonbonpony]

03:10 Where did the formula come from?

[@nathaliesldga911]

Hi. I believe you've made an error at 18:10 when you state that post-alveolar obstruents are [-distributed]. I think post-alveolars (as well as dentals) are always [+dist]; alveolars and retroflex are [-dist]. In this case, the rule should be that [+ant] becomes [-ant], and [-dist] becomes [+dist] in the environment mentioned. Please correct me if I'm wrong but this is what several academic sources reveal. Thanks, Nathalia

[@melfisa]

ha! right, it looks like I didn't pay attention to the lecture, sorry:) in my memory it was 1,2,7,8 high and 3,4,5,6 low. I rewatched it again and it turns out I'm arguing where I actually agree with what you say (just not with what I remember). It's still redundant in purely mathematical sense but it makes perfect semantic sense.

[@melfisa]

actually 4 features is enough to define all 16 values: 2 for height, 1 for roundness, 1 for frontness. but my question was why you use 3 features to define height alone with low always being opposite of high... is this explicitly redundant on purpose (like control bit)?

[@parlazzy_6158]

Excellent

[@meryemraoui8995]

Thx for the effort

[@larkinlover]

Fantastic, thank you!

[@stealthmd]

Incredible lecture! Thank You very much!

[@sighisoaraa]

The difference between a sonorant and a continuant is very subtle.

[@yourway382]

Do you still need an explanation?

[@tanumrita]

Thank you, professor.

[@msclrhd]

The problem I have with the distinctive feature representation is: 1/ it does not represent all IPA consonants/vowels (limited to English, no lax/tense vowels, trill, flap, click) 2/ classification does not appear to be standardized 3/ inconsistent modelling of features: +/- vs +/0/- vs + 4/ allophones and co-articulation not represented In the plural example you could have said sibilant fricatives => /Iz/, voiced => /z/ and voiceless => /s/ (given that as you mention, vowels are voiced).

[@ulfr-gunnarsson]

I think, it is still applicable for any language, but you need to define what features are distinctive for a given language. In this lecture (and its first part) there is an overview of distinctive features that can be used for any language.

[@user-zy3bo3rx8x]

thx a lot!!!! I love this video!!!! it's so enlightening!

[@akbarrana9115]

thank u very much sir. it's really really helpful

[@TheShadyblida]

great lecture !! thanks

[@walker6487]

i want to know how many distinctive features in phonology ?!

[@ngochongnguyen3379]

My idol 😍😍😍😍

[@abramoibra3857]

If you may Sir, can you explain to us in the coming videos the difference bitween the complimentary distribution and the free variation in phonology, thank you so much.

[@taqwaajannatiall7411]

Peace, Mister

[@luckyadams4041]

Thanks for trying

[@taqwaajannatiall7411]

أنا متواصل Mister معاك في كل الدروس لكن عندي طلب اريد ترجمة بالعربية اسفل الفيديو طبعا فضلا وليس امرا لان اللغه عندي ليس مئة فا المئة thank you very much