Many languages insert consonant sounds to keep vowel sounds apart. In my speech, I insert ‘r’, so I say drawring [d͡ʒɻoːɻəŋ], not drawing [d͡ʒɻoːəŋ], and lawr and order.
Glottal stops, [h], [ɻ], and glides are commonly used. However, [p], [k], [m], and [v] never are.
The same kind of asymmetries show up in many different aspects of phonology – and morphology and syntax, too.
A lot of my work has been about trying to figure out what the asymmetries are and how to explain them.
The challenge is that there can be many reasons for asymmetries, and not all involve cognition, and not all involve phonological cognition. For example, certain structures might be hard to learn for perceptual reasons. So, part of the challenge is teasing apart all the influences on asymmetries and finding which part is cognitive.
Markedness
The theory of asymmetries is often called ‘markedness’. I’ve written quite a lot about markedness, and I have changed my mind about what the phonology does over the past several years. The best place to start is here:
A free alternative is to read my doctoral dissertation – it’s about the same theory except the dissertation has a lot more interesting stuff (about assimilation, vowel reduction, and coalescence).
de Lacy, Paul (2002). The formal expression of markedness. Doctoral dissertation, University of Massachusetts, Amherst. Amherst, MA: GLSA Publications. ROA 542.
If you want to read something shorter:
de Lacy, Paul (2004). Markedness conflation in Optimality Theory. Phonology 21.2:145-199.
… and there’s also a free version:
And here’s a tutorial:
Next steps
Nowadays, I don’t think my early theory is on the right track. You can read about that in:
Positional Markedness
I also have a paper about ‘positional’ markedness, which talks about asymmetries in specific prosodic environments: