A great paper by Jonathan Bennett can be found here (thanks to Ektopos for the link). The title is "On Translating Locke, Berkeley, and Hume into English." I find it fits my experience entirely, alas; students just don't know how to read early modern philosophy, and I'm inclined to agree with Bennett's diagnosis: "Their schooling has been such that they have never learned properly to read anything; and the habits of impressionistic approximation which they picked up there are not seriously opposed in many departments of the University."
I have doubts about the extent to which the proposed solution really helps, however. That's not to say I think it without value, since I think it helpful to an extent. It would be very useful for high school philosophy classes, for instance, and probably, (used with discernment) for 100-level undergraduate classes, too. But I think there are far better ways to deal with the problem. I've been trying for some time to develop an adequate response to this problem in my own teaching. Since I haven't been teaching very long, I'm still relatively new at it all. But I find that the most effective thing is guided reading in class (of which there are several different variations, some of which work better than others). The major impediments to this are time and endurance. I'm currently teaching classes that are three hours long; we started guided reading in Berkeley's Three Dialogues (each student read a passage, gave their first impression of it, and I commented). The First Dialogue, which we didn't quite finish, took the whole three hour period, excepting a ten-minute break, and by the end the students were noticeably fatigued. (On the other hand, I covered most of what I would have covered had I done straight lecture, and I'm fairly sure that the students took away more that was really Berkeley himself than they would have done under most approaches.) Another variation is to have students paraphrase the text themselves, and then go over it with them. There probably is a better variation on the guided reading approach than those I've used so far, but I prefer this approach (admittedly more difficult for both teacher and students) because:
1) The whole "it's the language" complaint is the one type of student complaint for which I have no sympathy whatsoever; when students use that one on me, I reply, "So now you have the opportunity to learn real English." I don't have problem with updating spelling and (to a lesser extent) punctuation, any more than I have problems with updating typeface; beyond that, I am not inclined to budge on the issue. 18th-century English is not convoluted; it involves a more complex, less eighth-grade-level quasi-journalistic style than is common today, since we oversimplify our sentences, but all it takes to read it is a bit of practice and familiarity.
2) I'm inclined to think that the language is much less of an actual problem than it might sometimes seem. Some students do have difficulty with it - Bennett is right when it comes to ESL students, for instance, and some students are just atrocious readers. For the most part, however, I've found from my (admittedly limited) experience that they can read it when they expend the effort, and quite well (only occasional slip-ups due to the language); they just don't consider it a high priority. I have doubts about the wisdom of what might amount to accommodating people's refusal to consider philosophy important.
3) My goals in teaching are also, I expect, somewhat different from Bennett's. My primary interest is to try to give the students a feel for the actual text that they might carry forward; obviously, there's some tension between this and giving the students a modified text. I am less interested in arguments (Bennett's 'intellectual content') than with the classics with which a great student of early modern philosophy can wrestle the rest of his life through. I see no particular value in postponing this to graduate-level studies. Further, in the case of the Scottish writers, they spent an immense amount of time and effort working to write good 18th-century English. If they could have the courtesy to try to write 18th-century English, we can have the courtesy to try to read 18th-century English. The distance between 18th-century Scots English and 18th-century standard English was much greater than the distance between 18th-century standard English and our own standard English. But maybe this is just all a sign that I'm getting old-fashioned. Sigh. So it begins....
-> I'm a little puzzled, by the way, at Bennett's modification of Cottingham's translation of Descartes; while Cottingham's translation is more or less satisfactory, it seems to me that it would have been far better to have translated colloquially directly from the French. 'Translations' of translations leave me a bit skeptical; for instance, I think Bennett may be misled by Cottingham's translation; Descartes's original seems to me to be rather far from sarcasm. Compare the Cottingham translation Bennett gives with the more closely literal John Veitch translation (paragraph 5). But I'm not a Descartes scholar, so it could be that I'm missing something.
-> It took me this long to realize (tomfool that I am) that there's a spellcheck option on the post interface! Woohoo! You can all look forward to posts with fewer typographical errors....
