Bill Vallicella has a nice discussion, with his usual clarity, of MacIntyre's argument that evaluative conclusions can be drawn from factual premises. However, I think he makes a mistake, one easy to make in his reply, and it's an interesting one. He says (of the example, "This watch is inaccurate, therefore this is a bad watch"):
Speaking as someone who has been more influenced by the moderns than by the ancients, I don't see it. It is not the case that "the concept of a watch cannot be defined independently of the concept of a good watch . . . ." A watch is "a portable timepiece designed to be worn (as on the wrist) or carried in the pocket." (Merriam-Webster) This standard definition allows, as it should, for both good and bad watches. Note that if chronometric goodness, i.e., accuracy, were built into the definition of 'watch,' then no watch would ever need repair. Indeed, no watch could be repaired. For a watch needing repair would then not be a watch.
The assumption that is made here is that if the concept of a watch cannot be defined independently of the concept of a good watch, then being a good watch would have to be part of the concept of a watch. But this is surely not what MacIntyre means. Rather, the claim is that the definition of a watch contains or implies a standard according to which something can be a good watch. The point is not that chronometric goodness is built into the definition of a watch, so that all watches are good watches, but that you cannot understand something as a timepiece without relating it to standards for keeping time, and from the fact that a watch is a timepiece one can recognize that between two otherwise equal watches, if one is more chronometrically accurate, it is better qua watch.
MacIntyre focuses on the timepiece or time-keeping component of the concept, but in some ways the point is easier to see with the other component of the Merriam-Webster definition that implies evaluative standards, portability. Portability is something that admits of degrees. You could strap a timepiece the size of a small alarm clock to your wrist; it would be portable, but would obviously be less portable than a normal wristwatch. Portability is in fact also relative to your end, because it is a feature describing how well a means relates to certain kinds of ends. For one kind of end, 'portable' might mean you can drive it around in a van; it is pretty clear that 'timepiece you can move around using a van' is not an adequate standard for watch-portability. As the Merriam-Webster definition says, the portability of a watch is such as to be put in a pocket, or strapped to your wrist, or strapped to your ankle, or hung around your neck as a necklace (to take the most common cases). But even granted this, not all watches are equally portable. And if the definition of a watch includes portability, this means that not all watches will capture equally well this aspect of the definition. It is not a part of the definition of a watch that it excel at portability; but it does seem implied by the very fact that portability is part of the definition that one can evaluate a watch according to its portability. And if this is the case, then it does appear that you can conclude from a watch not being very portable that it is, to that extent, not very good as a watch. The same thing is true of the 'timepiece' component and (possibly, if it is not taken merely to specify the other two) the 'designed' component of the Merriam-Webster definition.
If this is true, does this mean that "Watches are portable timepieces" is an evaluative, not a factual statement? As far as I can see, it does not. The fact that a statement includes or implies the fact that there is a particular standard of evaluation, does not itself imply that the statement is an evaluative one. "Watches can be evaluated according to how well they keep time" is, despite explicitly mentioning evaluation, a factual statement, not an evaluative one; in making the statement, you are not evaluating watches but recognizing the fact that watches can be evaluated. It is, after all, a fact that things are evaluated; it is a fact that when they are evaluated, they are evaluated according to standards; and it is often a fact whether or not something has the features that make it fit those standards or not. Because they are used for measurement, and because they may be more or less useable for such measurements, watches are as a matter of fact things that can be evaluated according to how well they serve for such measurements. Means of measurement may be more or less good; they may even be bad; but it is not intelligible to say that something is a means of measurement and yet there is no way to assess whether it is good at measuring. But it is a fact that timepieces are things used for measurement, and it is a fact that they can be more or less good at measuring; and there are facts that will guarantee that it is less good.
One could, of course, go another direction and argue that all statements are in fact evaluative; that there is no such thing as a statement that is purely factual without anything whatsoever that is evaluative. But again, that a timepiece is evaluatable according to a standard of good time-keeping seems a fact about timepieces, not an evaluation of time pieces. It's not even normative; it's just true as a matter of fact that you can evaluate the quality of a timepiece according to its time-keeping. It doesn't make any sense to talk about timepieces without any regard for whether the things in question can be assessed according to their time-keeping ability. If you were going to do that, you might as well say that I can create a watch by strapping a rock or a living mouse to my wrist: "Look at my mouse watch, it's a portable timepiece, although I have no idea even what standard might be used to assess whether it is any good at keeping time." To say it is a timepiece already implies that it can be evaluated according to its time-keeping; this does not make the existence of timepieces any less matters of fact.