TeX - LaTeX Stack Exchange is a question and answer site for users of TeX, LaTeX, ConTeXt, and related typesetting systems. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

In Vertical space between section header and text is too big user egreg said that setting \predisplaypenalty=0 is wrong idea.

Why is that? Could you provide an example?

If not '0', what value could be used?

share|improve this question
up vote 11 down vote accepted

The parameter \predisplaypenalty is the "cost" that TeX assigns to inserting a page break immediately before the start of a displayed equation (or multiple equations). On p. 189 of the TeXbook, Knuth writes:

Plain TeX sets \predisplaypenalty=10000, because fine printers traditionally shun displayed formulas at the very top of a page."

[Asides: (i) If a TeX penalty parameter is set to 10,000, it's the functional equivalent of infinity. (ii) LaTeX also sets \predisplaypenalty=10000.] If one were to set \predisplaypenalty=0, this cost parameter would be 0, i.e., TeX wouldn't even try to find alternative page break points than the one it may have "found" immediately before the start of a displayed equation.

As with virtually all typographic "rules," this rule is not absolute. There may be some circumstances in which it's better to violate this rule than it is to violate some other, even more important typographic rule. Nevertheless, disregarding this typographic rule entirely -- by setting the cost parameter to 0 -- cannot be the best approach.

share|improve this answer
Why do you suppose it became a "rule" not to start a page with a displayed equation? My guess is that it's because it looks bad, to me, for a page to end with a line that doesn't go to the right margin and that doesn't end with a sentence-ending punctuation mark. For instance, if I see "Thus we see that [rest of line is blank]" at the bottom of page 305, I think for a moment that something got cut out, even if page 306 begins with the displayed equation that logically follows. But that's just my guess. – MSC Mar 14 '13 at 19:26
@MSC - In addition to your guess (which seems entirely plausible to me), I'd venture to guess that this "rule" also reflects the general strong aversion to (typographic!) "widows", i.e., starting a page with a single line before the next paragraph break occurs. Starting a page with a displayed equation can look very much like having a typographic widow. Cast in these terms, your conjecture might be (re)stated as saying that having a page break right before a displayed equation not only creates a widow on that page but also an "orphan" (or "semi-orphan"?) on the preceding page... – Mico Mar 14 '13 at 19:39
It all depends, I think. At my current project, I'm probably content with \predisplaypenalty=\@highpenalty. That allows for some pagebreaks above equations, where the preceding page would otherwise contain huge white gaps around equations or similar uglinesses. Short two-line paragraphs between equations are not beautiful either, when they get broken across the pagebreak: real widow and real orphan in one go. – Blackface May 20 '14 at 14:31

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.