Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Are there more sounds?

  1. Jul 4, 2013 #1
    We have given a symbol for every sound that we can produce. Are there more sounds? There is a possibility that there exists more types of sounds but since we cannot produce them we haven't given a symbol to it yet.
    Any thoughts?
  2. jcsd
  3. Jul 4, 2013 #2


    User Avatar
    2017 Award

    Staff: Mentor

    Who is "we", and which type of symbols do you mean?
    Human bodies can produce sounds which cannot be expressed in English.
    Our machines can produce even more sounds. They can basically generate every possible sound, as they can produce (nearly) arbitrary wave patterns.
  4. Jul 4, 2013 #3
    I'm a bit unsure of my question!
    so basically English has assigned alphabets only for a set of symbols?
    And can we or can we not represent every sound with English alphabets?

    "we" meant human beings by the way
  5. Jul 4, 2013 #4


    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper

    Different languages have different sounds and use different methods to represent those sounds. By 'sound', I am specifying only the sounds used in human speech.

    The English alphabet has 26 letters which either singly or in combination represent the sounds used when uttering English words. Not all languages use the English alphabet, with a,b,c, etc. Greek and Russian have their own distinctive alphabets, and these languages also use sounds which are not found in English.

    Not all languages use alphabets. Some languages, like Japanese, can be written with an alphabet, or they can use special signs which represent syllables (Japanese has two different scripts using syllables), or they can use a large set of abstract signs called ideograms, where each ideogram represents a different word and where there is no correspondence between the sound of the word and its ideogram. Japanese ideograms were, to some extent, borrowed from the Chinese, although Chinese and Japanese are unrelated languages, like French, Italian, and Spanish are.
  6. Aug 22, 2013 #5
    there are sounds like ultrasonic sound that animals like dogs and bats hear/make that we can't

    among languages there are also sounds that don't appear. For example, the rolling r sound is in spanish but not english, and there's also the example of the chinese not being able to say their l's, this is because they don't have the l sound in their language (so before the chinese encountered people who used l's it was like no l sound existed)
  7. Aug 22, 2013 #6


    User Avatar
    Gold Member

    A sound can most completely be characterized (without information loss as far as human perception is concerned) by a short (~1s) time series of pressure measurements made at ~44k Hz (any higher and it's mostly a "quality issue".. it's technically different and can technically be percieved differently, but probably not with any new qualitative features).. which maybe brings up the matter of amplitude. That can be arbitrary to a certain extent. It's the same qualitative sound, just louder or quieter (yes, there's pedantic exceptions at extreme volumes).

    There's countless combinations ("almost" infinite) that you can still describe this way, so you may as well just use the pressure-time series at this level of resolution, since no human alphabet could reasonably contain all the combinations.

    For many cases, the quality can be simplified even further, to describe the wave at a higher level in bigger parts: the attack, decay, sustain, and release.


    Now this is not even a unique reduction (there's still serveral higher order harmonics that we can perceive and produce with different vocal techniques) and we still have no alphabet large enough to contain the ADSR envelope possibilities.

    People who write synthesizer sounds (nonvocal now) have come to just give different sounds whole names that reflect what they sound like "aliens crying", "metal falling", or completely wacky names like "hallucinogen purple" because it's pointless to try and order it with a symbol system since the possibilities are so vast. You could technically do it, but then you'd have no idea what each symbol meant. If you want to make frequent use of a set of sounds in a structured way, you just have to pick a limiting set and assign it some symbols (like a culture's linguistic alphabet) or memorize a batch of qualitative descriptions you like (wet drums, deep bass, dobro guitar, lucifer yellow, etc).
  8. Aug 24, 2013 #7
    French, Italian and Spanish are unrelated? They all come from Latin. By understanding Spanish you can understand much of written Italian and Portuguese.
  9. Aug 24, 2013 #8


    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper

    My clumsy phrasing meant to say that, yes, the modern Romance languages derive from Latin, whereas Japanese and the various dialects of Chinese are not related and share no common ancestor language.
  10. Aug 25, 2013 #9


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Years ago my wife's grandfather, an old Bohemian, attempted to teach me Czech. I learned that Czech is one of the most modern languages having be completely codified very late on the scale of European languages, if you can say the alphabet, you can read.

    For Americans the trouble is saying the alphabet. One character in particular, written as an r with a hat is said as a combination rolled r and sh sound. He claimed that he had never encountered an American that could say it correctly. I never could say it. The point?

    The English language does NOT capture every possible sound.
  11. Aug 25, 2013 #10


    User Avatar
    Science Advisor
    Homework Helper

    Strange - I've never heard Brits having problems pronouncing the composer Antonín Dvořák.
  12. Aug 25, 2013 #11

    jim mcnamara

    User Avatar

    Staff: Mentor

    Just because a "sound" exists in the human vocal stream does not mean it maps to one character. There are 126 identified phonemes (as of 1986 when I knew this stuff). And in English single vowel sounds map to more than one character. Ex: bye, buy, by, "bi"nomial; be, bee.

    I've mentioned this concept before. In a previous post. It caused a lot of debate. Some folks here who believe in (IMO only) infinite non-ending neural plasticity of the human brain do not want to ascribe real meaning to this concept: there are phonemes in some languages that your brain "wiring" cannot reconstruct recognition of nor can it create the sound of, once the brain passes the age of about 1-2 years. This occurs when the sound is not part of extant speech. The perception of the sound vanishes. This is attributed to the fact that we are born with LOTS of possible synaptic connections - once lost, some synaptic connections cannot be reconstructed.

    And it is okay to dismiss that idea, but having decent research behind it would be required before I'll go along.
    I cannot find the original citation for the comments above - anymore.
    Non-native pronunciations of English - Wikipedia, the free encyclopedia

    Since you appear to be an English speaker, consider that you probably will have trouble hearing and producing some of the other 70+ phonemes English does not have. Researchers have worked very hard to categorize humans speech sounds. So they are probably close to having all of them, as a guess. ESL researchers work very hard on the particular subject: How do others hear English?

    So there is a "sort-of" answer to your question: orthography comes only after you can hear the sound. In my example above the researchers could "see" what was going on, using digital frequency analysis, but they could not hear it. Nor could they produce the sound so native speakers understood it. So what do you do? Create a new phonological symbol. And hope one of the native speakers gets an advanced degree in linguistics.
  13. Sep 24, 2013 #12
    You mean language? English has many more sounds than symbols.

    If you mean sounds in general there is a huge number of them, very few of which have symbols.
  14. Sep 24, 2013 #13
    It's actually the Roman alphabet, which is used to transliterate English. This is one reason spelling English is strange, with the a in tape and tap pronounced differently. Why wasn't a separate symbol used for the two distinct sounds? I doubt anyone knows. Many other languages are transliterated to Roman letters, but often accent marks are added.
  15. Sep 25, 2013 #14


    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper

    The English alphabet is based on the Latin alphabet to be sure, but English has added letters which were not used at all by the Romans or were used only to write words from foreign (to the Romans) languages. (J, V, W, and Y in particular were added after Rome fell. K was rare (Kalends), as was Z, which was used for certain place names or for words borrowed from Greek)

    Some language scripts (like French and Spanish) use diacritical marks or accents to indicate which syllable is stressed or how a vowel is to be sounded. Greek uses different letters sometimes to indicate a long vowel sound (omega, literally 'big o' v. omicron, lit. 'little o'), but in general, the Indo-European languages use different means to indicate the length of the vowel sound instead of adding letters. This keeps the number of letters to a minimum.

    Some languages, like German, will use a double letter to indicate a lengthened vowel sound, like the word 'Meer' (sea). German also will add an 'h' after a vowel also to indicate the vowel is long, e.g. 'Sohn' (son).

    In English, the rules are more complicated. The word 'tape' falls into a category where a single-syllable word ending in 'e' usually has a lengthened vowel preceding the 'e', so we have 'tape' v. 'tap' or 'nape' v. 'nap', 'hope' v. 'hop', 'gape' v. 'gap', and so on. This category has exceptions like 'pare', where the 'a' is short, but 'pane', where the 'a' is long.

    One of the reasons why English orthography (how a language is written) is complicated and contradictory is that printing caused the written language to become standardized while great shifts in pronunciation were occurring or about to occur. For English, printing in the 16th century caused the language to be written in a certain way, but in the 17th century, a 'Great Shift' occurred in how some vowels were sounded, but this shift was not reflected in major changes in English spelling. In the great works of English literature written in the 16th century, like Shakespeare's plays and poetry, there are a number of rhymes and puns which were used and understood by his audiences which are no longer recognized by modern audiences because of this shift in English pronunciation.
  16. Dec 13, 2013 #15
    I've studied Chinese for 5 years, and they do have L sounds.

    The problem as it were is with the English L, not the Chinese L. Of course, if you are a native speaker of English, you might not realize that, but the English L is a particular kind of L, it is 'dark' unlike in Chinese and many other languages, and to them it almost sounds like an N (even if to you, again, the difference might sound cristal clear). And this, in turn, is because the Chinese N is much softer and discreet than the usual 'N' found in other languages including English.
    For instance, most Chinese have absolutely no problem pronouncing a French, Spanish or Italian L.

  17. Jan 27, 2014 #16
    There are some Asian languages (Chinese, Vietnamese, etc) that have many sounds with no English equivalent. I am told that the sound "who" has many variations in these Asian languages.
  18. Feb 11, 2014 #17
    for example, in Japanese the syllable ふ is somewhere between fu and hu. The 'f's in futon or coffee for example don't really sound much like how English speakers pronounce them (palatal instead of glottal).
    Last edited: Feb 11, 2014
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook