Style and tendency
In explaining musical styles, Leonard Meyer divides musical characteristics into three categories: laws, rules, and strategies. Laws are characteristics of music that are based on human biology and psychology, and as a result laws are more-or-less universal. Rules are culturally conditioned. They are hallmarks of a particular style that are more-or-less universal within the style, but differ from style to style and culture to culture. Finally, strategies are specific ways in which composers work within a style — the things that make one composer's work sound different from another's, even if they compose in the same style.
For the most part, principles of voice-leading or harmonic progression are "rules" according to Meyer's definitions. They are specific to a style. Or, in some cases, they are shared among a few styles of Western music, but are far from universal. Thus, it can be helpful to think of them as collective traits of some music(s) we seek to understand and emulate, rather than hard-and-fast it-must-be-done-this-way strictures for all musical practice.
However, these rules are also related to laws, in as much as they represent one set of practices that mediate the various demands on music from basic principles of human auditory perception and cognition. For instance, the prohibition against parallel fifths is a specific way in which Western tonal composers have mediated the conflict between tonal fusion, goal-directed motion, and independence of line. There are many other similar cases.
Note, however, that while "avoid parallel fifths" takes on the form of what we consider "rules" in day-to-day speech, Meyer's rules of musical style are different. Meyer's rules are descriptive: these things tend to happen universally, frequently, rarely, never, in specific situations. "Avoid parallel fifths" is a prescriptive instruction based on that descriptive observation: because parallel fifths occur rarely in this style, and only in specific cases, avoid them in your own strict-style compositions until we have a chance to engage those specific cases — all the while remembering that other styles may have different tendencies.
That word tendencies is an important one. There are rarely absolutes in the musical parameters we engage most as performers, analysts, composers, listeners, etc. The absolutes of common styles fell into our unconscious background long ago. Instead, what makes each piece, composer, or style special and unique — what we care about — are the little ways in which they bend the "rules," the ways in which they express, thwart, and play with the tendencies of the style they engage — the ways they play with our expectations as listeners.
Over time, as we familiarize ourselves with a musical style, the tendencies of the style become expectations in our mind, and composers can, in turn, compose with those listener expectations in mind. Though those tendencies are subjective, and to a large extent statistical, the shared stylistic knowledge and the shared psychological expectancy create a kind of quasi-objective language. That musical "language," like spoken/written languages, is both reliable and bendable/breakable. The meaningfulness of a piece of music is dependent on that reliability. But its specialness is dependent on the ability for the "rules" of that language to be bent, even broken.
With this in mind, as we progress in our study of voice-leading, we will encounter more exceptions to the prescriptive rules, even in strict-style composition, and our hard-and-fast strictures will transition more and more into the language of tendency.