Prefix and suffix affixation are the two most common affix types, attaching to the beginning or end of roots or stems. Affixes interact with stress patterns, phonotactic constraints, and semantic/grammatical properties of bases. Prefixes and suffixes differ in productivity, multiple attachment, and semantic transparency across and within languages.
From your work on morpheme types and morphological structure, you know that words are built from smaller meaningful units — roots, stems, and bound morphemes — and that these units combine according to structural rules. Prefix and suffix affixation is the dominant mechanism by which English (and most languages with derivational morphology) builds new words: attaching bound morphemes to the left edge (prefix) or right edge (suffix) of a root or stem. The asymmetry between these two affix positions is more significant than it might first appear.
In English, suffixes are far more powerful morphologically because they typically change the grammatical category of the base. Adding *-ness* turns the adjective *dark* into the noun *darkness*; adding *-ify* turns the noun *terror* into the verb *terrify*; adding *-ly* turns adjectives into adverbs. This category-changing property means suffixes do most of the syntactic work in derivation — they determine what slot the new word fills in a sentence. Prefixes, by contrast, usually preserve the category of the base while shifting its meaning. *Un-* added to *happy* (adjective) yields *unhappy* (still an adjective); *re-* added to *write* (verb) yields *rewrite* (still a verb). There are exceptions, but the asymmetry holds broadly: suffixes change category, prefixes change meaning.
This asymmetry has consequences for multiple attachment — the stacking of affixes. Because suffixes change category, they interact with each other in constrained ways: you must build words from inside out, and each suffix attaches to a specific category. Consider *unhappiness*: the structure is [[un-[happy]]-ness], not *[un-[[happy]-ness]]*. The suffix *-ness* attaches to adjectives, so it must wait until *unhappy* (an adjective) is formed; then *-ness* attaches to produce the noun. Prefixes, being category-neutral, can stack more freely, but they still interact with the phonological and semantic properties of whatever they attach to.
A central property distinguishing affixes is productivity — the degree to which an affix actively generates new words in the current language. Some suffixes are highly productive: *-er* (agentive, "one who Xs") applies to almost any verb to form a noun (*writer*, *runner*, *compiler*). Others are frozen: the *-th* in *warmth* and *length* survives in inherited words but doesn't generate new forms (*coolth* is a nonce coinage, not a natural derivation). Productivity interacts with phonotactic constraints — the sound-shape requirements of the affix and the base. The suffix *-ity* strongly prefers Latinate bases (*probability*, *felicity*) and sounds awkward with Germanic ones (*\*hardity*), while *-ness* is productive across both (*hardness*, *brightness*). These selectional restrictions are not arbitrary; they reflect the etymological strata of English vocabulary and the phonological contexts each affix originally evolved in.