At Shimer College, we pride ourselves on teaching “primary texts” as much as possible. The goal here is to make sure that all the reading we assign is “discussable,” which secondary sources providing background usually are not. In general, I prefer this approach, but so far in my Islam class, I have begun to see its limitations. Class discussions have a greater number of uncomfortable silences than usual, and it seems to be mostly because the students don’t feel equipped to approach the texts.
On one level, this is strange, because although Islamic intellectual traditions are of course different from their various “Western” counterparts, they’re not that different. Indeed, in many cases they are drawing from literally the same broad traditions as the “Western” sources we discuss in other classes — above all, the biblical prophetic tradition and Aristotelian philosophy. In the grand scheme of things, the surah “The Cow” from the Qur’an is more similar to Deuteronomy than different, and ibn Khaldun’s political philosophy is more similar to Machiavelli and Hobbes than different. Yet in both cases, the latter would most likely produce a much more fluid and comfortable conversation.
I’m not sure the problem is that the students lack background. It’s not as though the Islamic world has radically unintelligible social standards, and I’m confident that the texts I’ve provided have given them at least as good a rough and ready background on the general shape of cultural life as they tend to feel like they have of Greek society or medieval times. Most of their assumptions about the Greeks and medievals probably wouldn’t stand up to serious scrutiny, and I do sometimes cringe when students pull out a facile argument about how something would’ve made sense “in the culture of the time” — but that sense of initial familiarity, even if partly unfounded, gives them confidence.
Yet it’s not simply foreignness as such that’s at issue. I don’t have direct evidence here, but I suspect that students would be more prone to jump right in with texts from other “Eastern” traditions. They would come to those texts expecting to find fascinating new ideas unparalleled in the boring Western traditions with which they’re familiar — and even if their sense of knowing what to expect is unfounded, it gives them the confidence to get started.
With Islam, by contrast, American culture conveys a consistent impression of inscrutable Otherness, utter impenetrability. Orientalist scholarship of course exacerbates this, as even introductory works throw an unconscionable number of Arabic terms at their readers, creating the impression that Muslims make use of bizarre, foreign concepts that we Westerners can’t fully understand. Further, engaging with Islam feels much more fraught and dangerous than engaging with Buddhism. Political circumstances give us the sense that it’s urgently important to “understand Islam,” but fear of making insensitive remarks or perpetuating stereotypes produces a unique degree of reticence and caution.
In the short term, I plan to deal with this problem next semester by simply giving up and using more secondary sources in my course over the Qur’an, for instance by opening the course with Sells’ much-recommended Approaching the Qur’an and using a translation of the Qur’an with much more robust explanatory notes. If students feel like they need more background to get started discussing, I’m probably not going to achieve much by repeatedly insisting they don’t.
In the long term, though, I think the only solution for Shimer College, if it wants to remain faithful to its general approach while doing the needed work of exposing students to Islamic thought, is to include Islamic sources alongside Western sources as a self-evident part of the dialogue that we’re trying to create. This doesn’t mean that every course must include Islamic sources, but I do think we would be well served to include Islamic texts beyond the obvious realm of philosophy and theology — why can’t ibn Khaldun appear alongside Hobbes and Machiavelli, or why couldn’t we read portions of ibn Rushd’s commentary on Plato’s Republic? Why can’t we read Hallaq alongside Foucault?
Yes, this would mean cutting something out in order to make room, but maybe every “Great Books”-style curriculum would do well to give up on the self-defeating goal of “covering everything” and think more intently about what we want students to get out of the texts and their juxtaposition (aside from checking them off the list of books that “everyone must read”). I’m sure that if we really had to justify every choice on a pedagogical level, we would find that no, we don’t actually need to read every single canonical early-modern political theorist, for instance, since our students are not at the level where they can detect the fine distinctions among them anyway. And in some cases, we may indeed find that the Islamic sources are actually more accessible and more suited to a background-free, “Great Books”-style approach than some of the Western texts we throw at them.
An und für sich 
For what it’s worth, I’d stick to the primary texts, especially because they cause discomfort, while using secondary sources and lecture to provide historical context. Talmud and Zohar are no less impenetrable, but less politically fraught. Regarding Islam, I’d encourage the students to consider their own discomfort.
Every day’s reading will definitely include passages from the Qur’an and many will have other ancient/medieval (whatever you call pre-modern Islamic) sources as well.