The Art of the American Folk Preacher (Rosenberg 1970) analyzed folk preaching to se if it was different in its oral composition from the Balkan oral poetry analyzed by Parry and Lord. However, Rosenberg says, the American folk preaching may well illustrate some relatively universal features of oral composition (Rosenberg 1986, 695). Rosenberg cites Henry Fearson’s 1818 descriptions of African-American congregations to suggest that folk preaching has not changed much in its content or style over about 150 years. He classifies the sermon as having oral spontaneous composition but drawing on prior experience and exposures rather like what happens in a jazz jam session (Rosenberg 1986, 697). Elements of Scripture and also of songs emerge in the extemporaneous preaching. Rosenberg cites numerous examples of song lyrics commenting on biblical passages, even if they seem at least partly at odds with the details of Scripture (Rosenberg 1986, 698).
The basic structure Rosenberg finds in use is very simple. He calls it “text-context-application,” in which the preacher cites or reads the text, describes its setting, then spends the rest of the “performance” describing application to life (Rosenberg 1986, 699). Rosenberg says that the content and structure are very fluid, largely because the preachers believe they are serving as conduits of the divine, so they are not actually in charge of the content. Contrary to this idea, he cites Rosenberg and Smith (1975), who analyzed semantic clusters in several sermons (Rosenberg 1986, 700). Some sermons were constructed with themes in parallel to one another, then by pulling the themes together for one conclusion. Some sermons seem more like free association of themes. Others had clusters of related themes, though the clusters may not be closely related to one another (Rosenberg 1986, 701). This pattern, again, Rosenberg finds as similar to what Albert Lord observed as early as 1960 in Yugoslavian oral poetry.
Rosenberg asks whether, as Albert Lord assumed, the compositional ideas came about during composition or, as psycholinguists assume, that the idea preceded the utterance (Rosenberg 1986, 703). He finds that, at least in some instances, the key words seem to have been arranged in advance to create an analogy. In essence, the folk preacher seems to have a set of contextual formulae which can be pulled into action on demand, creating infinite possible outcomes (Rosenberg 1986, 704). Memory of the elements is very important to the process, but the contextual memory may be so acculturated that it is not used consciously. The syntax of the oral clusters is used to associate one with another (Rosenberg 1986, 705). The associations can be used to tie very extensive, complex, and lengthy ideas together. For this reason, even very complex sermons, when used repeatedly, tend to become more consistent in their structure. The ideas seem to solidify with repetition (Rosenberg 1986, 707).
The audience, or congregation, also brings memory and expectation to the events. Rosenberg observes that “[t]he congregation enjoys the sermon because they know what is coming next, and how it will be expressed” (Rosenberg 1986, 708). He notes that at times the congregation responses actually anticipate the preacher’s message. Rosenberg cites studies in which people understand even unintelligible sentences to mean what they assume to be the message (Rosenberg 1986, 709). This corresponds to a congregation expressing loud approval for a message which is only beginning to be spoken.
Rosenberg goes on to discuss grammar and rhythm of folk sermons. After making statements which seem to wonder at people’s understanding of non-standard English grammar (Rosenberg 1986, 709), he goes on to say that metrical lines in a sermon will not be precisely the same as those lines in normal conversation. They tend to have stresses and pauses at conceptual junctures, especially at the ends of parallel statements (Rosenberg 1986, 710). Syntax tends to remain simple, as spontaneous oral composition does not lend itself to creating complex syntax. More complex syntax is also more difficult to hear and understand (Rosenberg 1986, 711).
Rosenberg recognizes that folk sermons typically reach a point where they become metrical and chanted messages. The aesthetic is carefully planned and is developed further as the congregation responds or does not respond (Rosenberg 1986, 712). One of the primary goals which Rosenberg identifies is for the preacher to move the congregation to an emotional state where they “are hardly aware of the preacher at all” but moving to a state of catharsis (Rosenberg 1986, 714). Rosenberg illustrates the concept using extensive quotations and analysis of a sermon about Israel crossing the Red Sea. The shape of peaks and valleys all leading eventually to the climax of the sermon is clear. Rosenberg also describes metrical repetition combined with an alliterative pattern which he suggests is as important to the message as any of the semantics of the message (Rosenberg 1986, 718).
Rosenberg points out that the motivating feature of folk preaching is in may cases the rhythm and musicality which is generally invisible on paper. However, he goes on to cite two different speeches of Martin Luther King Jr., one which is much like an orderly, logic-driven speech and one much more like a folk sermon (Rosenberg 1986, 720-722). While King was able to engage in either kind of oratory, he would read the need of the occasion and audience to adjust his presentation. Other speakers from the tradition of folk preaching continue to show similar skills. Rosenberg cites at length Jesse Jackson’s speech from the 1984 Democratic National Convention (Rosenberg 1986, 724). Rosenberg’s conclusion is that the folk sermon style can be and is adapted to different uses, including political and community speeches, using precisely the same forms and compositional methods which are used in church settings.