John McWhorter has an interesting discussion of why some languages lose their inflections and begin depending on word order and function words, from a global perspective an unusual development:
I ask these questions [why Native American, Australian, and Bantu languages have kept their "bristling paradigms of prefixes and suffixes"] because my research increasingly suggests to me that for a language to shed its inflections, rather than consistently replace or even retain them, is less business as usual than the unexpected case. From a global perspective, languages appear to usually do this as the result of widespread acquisition by adults, whose ossified language organs tend to clear away languages’ “junk.”
Thus the inflection-shy nature of Romance compared to Latin—and Romanian’s remnants of case-marking are dishwater compared to Polish, Greek or Lithuanian—was due to imperfect renditions of that language being passed on in the context of invasion and imposition from the outside. English is the only Indo-European language in Europe with no gender marking on articles or nouns—ever notice that?—because of Vikings’ approximation of Old English starting in the eighth century. It is presumably no accident that Persian, with its low inflection and gender-neutral third person pronoun, has been lingua franca par excellence throughout much of its history.
I’m pretty sure I’ve seen this argument before, but McWhorter puts it well, and it’s one of the few supposed historical explanations for language change that makes sense to me.