Before I get into the promised geeqery, let me point out two things. First, I intend to learn Torah tonight. Shapiro doesn't find any rabbinic support for the custom of not learning Torah on Christmas Eve, and he questions the antiquity of the custom. Second, the point of all this is going to be that even if you believe in gematria, you should examine the statements that are claimed to be gematriatically equivalent, even if they add up.
Anyhow, on p. 326 Shapiro cites R' Moses Grunwald, Arugat ha-Bosem, vol. 2 (Brooklyn, 1959), p 146a, as claiming that
are gematriatically equivalent. I won't explain the phrases, because the explanations will require explanations, and you'd be better off reading Shapiro's paper. And my topic here is the gematriatical equivalency, not the content. The first of the phrases contains an abbreviation for "Hashem." If we use the gematria for the Tetragrammaton (as I always call it) instead of that of the abbreviation, both phrases do indeed have a gematriatical value of 2,659. If I got the gematriot wrong, please comment.
My Hebrew is minimal, but I'm pretty sure both phrases are incorrect in ways that render the claim of equivalency incorrect.
The final word of the first phrase is either an unusual but correct way of spelling the word for "your Torahs [plural]" or an incorrect way of spelling "your Torah [singular]." To me, the incorrectly spelled singular seems more likely. If it is the incorrect spelling, it's understandable--at the end of a phrase, "toratekha" is pronounced "toratekha" (with the underlining indicating the accented vowel). This -ekha ending is used in the plural and takes a yod, but the yod isn't used in the singular. If this were meant to be the plural "torotekha," it would usually take a vav after the resh, although it isn't strictly required. But it's unusual without it. And what would "your Torahs" mean anyway? It might make sense--the written and the oral Torahs--but it just rings oddly to me. On the other hand, I'm new in this neighborhood.
If I'm correct and the singular was intended, we can fix the phrase by deleting the yod, which would reduce the gematriatical value to 2,649.
In the second phrase "sha'ah" ("hour") is a feminine noun. Two of the other words referring to it, "zo" and "nofelet," are also feminine. But "bo" is masculine.
If "bo" is incorrect, we can fix it by changing it to "bah." The change causes a net loss of 1, making the corrected gematria 2,658.
In short, if I'm right that either phrase is incorrect, the gematriatical equivalency is incorrect. Put another way, the equivalency depends on incorrectness--it fails if the phrases are corrected. If I'm right about all this stuff, that is. Even if you believe in gematria, you may have reason to reject this particular equivalency.
More generally, even if you believe that something works, you should still read instances of it carefully. This is the real topic of my sermon.