Should I always compare something with/to something? In many texts you see structures like this one, from Maciej Paszynski, Fast Solvers for Mesh-Based Computations:
After several steps, we end up in the root tree, compare Figure 8.18.
Is it correct to say "compare Figure 8.18"? My understanding of the word compare is that you always compare something to/with something. I think it is more correct to say "see Figure 8.18," or "check Figure 8.18." What do you think?
Best Answer
I think what you see in the cited book's examples is a spelled-out form of cf., meaning (in Latin) confer, or in English compare. As far as I know, the abbreviation cf., when used in cross references or other citations, is never followed by with or to. But you're right that the same is not true of compare in a normal English sentence, where we might say something like "compare this result to the situation illustrated in Figure 8.18."
As a stylistic matter, I much prefer the cross-reference style "see Figure 8.18" to the style "compare Figure 8.18"—in large part because in many instances Figure 8.18 (or whatever) simply illustrates the point being made in the text; it doesn't contrast with it in any meaningful way—so what exactly is the reader being asked to compare the figure to?
Nevertheless, "compare Figure 8.18" is an acceptable and recognizable option to use (without with or to) in short-form cross references. The fact that the author refers to some figures in the cited book with the style "see Figure 8.23" and to others in the style "compare Figure 8.18) suggests that he may be trying to distinguish between instances where the reader should simply refer to the cited figure and instances where the reader should compare the cited figure with what was said in the preceding text. (Another possibility, of course, is that the author switches between "see Figure X.X" and "compare figure X.X" not to indicate a difference in how the reader should approach the figure, but from a sense that varying the word choice makes it more elegant, which it doesn't.)
As an aside, I note that the author of Fast Solvers for Mesh-Based Computations precedes each of his "compare Figure X.X" cross references with one of three punctuation marks on different occasions—a semicolon five times, a comma three times, and a colon once—though he could have used a semicolon each time. With similar inconsistency he uses "compare Figures 9.8–9.9" in one place and "compare Figure 8.16 and 8.17" in another. These are not signs of a well-copyedited book.