All languages display some form of the hierarchy of projections, to the extent we understand what this is: in a given clause, roughly, complementizers are higher than inflectional heads are higher than verbal heads, etc, modulo various kinds of syntactic movement. We can also talk about the ordering of determiners, numerals, and "quality" adjectives within the DP. Has anyone in the linguistics or philosophical literature written explicitly on why we see this pattern displayed again and again? Does it emerge from the nature of the syntax-to-conceptual system mapping (whatever that might be like), or is it (somehow) imposed by the language system? One can find detailed cartographic approaches like that in Peter Svenonius' Spatial P in English which are highly suggestive of the semantic nature of the hierarchy (perhaps suggesting it is not imposed by language?), but I can't find an extended consideration of its origins.
I have found a paper that addresses this question directly (finally!). Svenonius & Ramchand's 2014 paper (here) offers an explanation for universal "grammatical zones" that appeals both to innate grammatical principles and properties of extralinguistic cognition. From the abstract:
...there remains an irreduceable universal functional hierarchy, for example that which orders epistemic modality and tense over root modality and aspect, and that which orders the latter over argument structure and Aktionsart (as discussed in much previous work). This residual core functional hierarchy (CFH) is unexplained so far by work which follows MUG ["a minimal role for UG']. Rather than simply stipulating the CFH as part of UG, we reconcile CFH with MUG by detailing what nonlinguistic cognition must look like in order for MUG to derive the CFH. We furthermore show how an individual language develops a language- specific RFH which is consistent with the universal CFH, illustrating with a detailed account of the English auxiliary system.
No serious attempt has ever been made to justify in semantic (and ultimately ontological) terms the extremely rich inventory of functional categories that Cinque has been elaborating since the late 1990’s for both the clause and the DP (and the categories of their complements, i.e., the ‘entities’ that e.g., Cinque’s adverbials must predicate higher order properties of!), and I’m afraid, for the time being (and for many years to come, at best) you may as well abandon all hope of seeing most of the functional categories posited in his cartographic work grounded on empirically credible ontological/semantic categories.
Ernst’s 2002 book, Jackendoff’s much earlier (1972) one, Ross's and other 'generative semanticists' work in the late 60's and early 70's, Beth Levin's work, Abney's dissertation, and a few other works (including those in other linguistic traditions, especially Fregean-Montagovian and Categorial Grammar, but also Dik’s and Rijhoff’s FG work on the structural levels in the clause and the NP, respectively, and Hengeveld’s recent work on Functional Discourse Grammar, however sloppy its 'LF' may be) have contributed to some extent to ground a few of those functional categories, but most remain purely stipulative (actually, most of the base-generated categories a Cinquean analysis entails do not even have an adequate descriptive name, not to mention the innumerable additional and completely ad hoc categories that must be posited if observed surface order is to be derived from a Kayne/Cinque-style account of base-generated structure).
Although ontologists as austere as Quine (or the less austere, but still austere Davidson) may not be in business any more, I suspect, not even the most extravagant contemporary philosophical ontologist would subscribe to the kind of ontology that a Cinquean theory of Language would call for if his syntactic categories were to be invested with credible semantic/ontological content. Hence, although the empirical evidence that has been brought to bear by Cinque and his followers on the existence of such extremely articulated and scope-wise rigid hierarchies inside both the clause and the DP is impressive, all that - in itself excellent - work will have to remain empirically adequate (i.e., predictive enough) perhaps, but merely stipulative, ontologically/semantically vacuous, and unable to really 'explain' anything, at least until linguists decide to investigate intra-linguistic ontology (as opposed to extra-linguistic ontology) in depth.
Unfortunately, as far as I know, no sustained work has been done on intra-linguistic ontology since ‘structural semantics’ and particularly Coseriu’s semantic programme became extinct, or at least confined to very obscure academic quarters, in the early 1980’s. The early Chomskian programme(s) (e.g., Aspects, Language and Mind, Reflections on Language, Knowledge of Language), of course, entailed the elaboration of a proper linguistic ontology as a basis for the semantic component of the Language Faculty, but ever since the mid 1970’s the Chomskian programme has given up on that crucial goal, and, particularly since the Minimalist Programme was launched in the early 1990’s, the content of the Language Faculty has been quickly contracting into triviality. Once the lexicon is factored out, what remains is basically Merge (recursion); practically everything else has now been irresponsibly declared the psychologists’ or the cognitive scientists’ job, and, of course, such specialists, like all, or the great majority of, philosophical ontologists, do not believe in an independent linguistic ontology.
As a consequence, nobody knows or, I'm afraid, is even likely to ever find out, what entities correspond to most of the Cinquean categories. Individuals (> DPs), sets (> unsaturated predicates), events (> saturated VPs, or hidden components thereof), states of affairs (> predications/IPs), modal predications (> ModPs), maybe propositions (> Polarity Phrases), and speech acts (> ForcePs) may be ontologically justified (if your ontologist is very generous), and it is relatively easy to justify their respective scope, but that’s about it, as far as I can see.
That’s what makes life so hard for any syntax and semantics teacher that seriously tries to convince his students of the existence of a transparent, compositionality-based, relation between minimalist syntactic representations and the 'semantic' representations the C-I systems are supposed to derive from them at the conceptual-semantic interface.