Recently, a video has gone viral of a young Finnish woman speaking gibberish that sounds like what certain languages sound like to foreigners. If you watch the video below, you can see that although some of them aren’t perfect, many of the languages she imitates sound remarkably realistic.
How is it that you can make yourself sound like you’re speaking a language, when you aren’t using any of the words of that language? The bulk of this type of performance is based on phonology, the patterning of sounds in a language. First of all, a Youtuber must know all or most of the sounds of the language they are imitating. If they choose to imitate English, they must know that English contains several “unusual” sounds, like our “r.” To make the gibberish really sound like English (as opposed to another language), they should use that sound a lot — the more usage of those distinctive sounds, the better.
Not only does the Youtuber have to know the inventory of sounds for a language, but they must also know what sounds can go together. For example, in English there are many sounds you cannot use in combination, either at the beginning of a word, the middle of a word, or the end of a word. You cannot start a word with a combination of the sound “k” (as in “car”) and the sound “th” (as in “three”). So a word like “kthar” is not acceptable, even as a gibberish word for English. The sounds “m” (as in “mother”) and “p” (as in “police”) cannot occur at the beginning of a word like “mpal”, but are considered totally fine in words like “bump” or “clump.” Some combinations cannot occur anywhere, such as “s” (as in “sun”) followed by “b” (as in “bump”); therefore, “tisb” would not be a good word to use. A word like “beemp” or “snorl”, however, would be perfectly fine.
In Japanese, where the phonological rules are different, a word like “beemp” would not be okay, even though Japanese also has the sounds “b”, “ee”, “m” and “p.” Syllables of Japanese cannot end in a consonant cluster like that. To imitate a Japanese word, you’d need to use what linguists call “open syllables,” which are syllables that do not end in a consonant. The only consonants that are allowed to end a syllable in Japanese are “m” and “n”. Think of every Japanese word you know; they will all follow this rule. Sudoku, karate, samurai, sensei. In addition, Japanese words do not allow consonant clusters like English does. You can see how this plays out in English loanwords in Japanese. A Japanese word for a labor strike, from English strike, becomes sutoraiki, inserting vowels so that it becomes a “good” Japanese word. So a good fake Japanese word would be something like “panaku,” whereas a word like “stranum” would not be a good word for Japanese, but would be perfectly fine in English.
I am not suggesting that the people who do these videos know about all these steps consciously. I would guess that in reality, they learn to do the fake language by doing what we all do naturally, as language learners: listen and imitate. Even professional linguists sometimes have trouble determining what phonological rules are at play in a language. However, in order to put on a convincing linguistic performance, a rudimentary knowledge of phonological rules for a given language is necessary.