6

I heard someone say this today, and it dawned on me that I cannot see a clear source to it.

Does this go back to adding machines and other mechanical integrators? It seems plausible, but the term does not appear in print until around the mid 1970's.

update: Per @JHCL's comment: "number-grinding", and related forms, maybe very relevant to this topic — but they do not yet appear in Google ngrams. Great digging thus far...

enter image description here

Improve this question
12
  • 5
    Because numbers are so crunchy. Words, on the other hand, are chewy. (I would guess it goes back, at least in part, to the sounds made by mechanical adding machines -- sickly green Marchants.)
    – Hot Licks
    Aug 20, 2016 at 23:24
  • 1
    books.google.com/ngrams/…
    – Hot Licks
    Aug 21, 2016 at 1:25
  • 1
    @Matsmath - I suspect what happened is that the concept of a dedicated "number cruncher" lost traction in favor of "general-purpose" computers that could handle the number crunching task.
    – Hot Licks
    Aug 21, 2016 at 1:40
  • 1
    @Matsmath - In particular, it was around 1985 that it became common for all computers to be equipped with high-speed floating-point hardware, vs having that as an option that was only present on "number crunchers".
    – Hot Licks
    Aug 21, 2016 at 3:03
  • 3
    Of course. It doesn't sound like grinding to me. It does sound like scratching, as evidenced by the quote I posted. If it had said 'chalk-grinding' I might have agreed, but the word 'number' makes me think it's a metaphor for calculations, not physical marks.
    – JHCL
    Aug 23, 2016 at 12:28

1 Answer 1

Reset to default
1

According to the The American Heritage Dictionary of Idioms, 2013:

crunch numbers: Perform numerous calculations or process a large amount of numerical data. For example, Preparing John’s presentation to the Federal Reserve Board required many hours of crunching numbers. This term originated with the computer age and indeed still applies mostly to the operations of computers. [Slang; second half of 1900s]

This could be related to the mechanical conception of punch cards for early computers, and the sound made by the machine while computing.

For instance, in Law Enforcement in the United States, James A. Conser, Gregory D. Russell, Terry E. Gingerich, Rebecca Paynich, Appendix 11-A, The evolution of information technology, one finds:

1943 : IBM and Howard Aiken succeeded in building the Mark I, a machine using punched paper tape that could "crunch" numbers up to 23 digits long.

A reference to crunching also appears on the above link to H. Aiken:

he saw how fields like mathematics and even sociology were hampered by their ability to crunch numbers—not unlike today’s push for taming Big Data.

In Makin' Numbers: Howard Aiken and the Computer, 1999, Cohen et al., one read that many cycles of operations could be running without visible mechanical motion,

however, even in this event the trained ear would know that the machine was still "crunching" numbers

Apparently, crunch came around 1795-1805 as a blend of craunch and crush.

Improve this answer
3
  • The example with the dateline "1943" is not (as in a citation-based dictionary) the date of first appearance of the sentence that follows. Rather it is the date when the thing described in the sentence happened. The sentence (and hence the use of crunch in it) is thus attributable only to the authors of Law Enforcement in the United States, third edition (2013). The linked article about Howard Aiken likewise provides no textual citation of an early use of "number crunching." So for the moment, AHDI's very rough estimate—"second half of the 1900s"—appears to be the best we can do.
    – Sven Yargs
    Aug 25, 2016 at 3:23
  • @Sven Yargs I am still looking for by precise dates. However I understood the question as "Why do we crunch?" and not "When did we crunch first?" I hope I have provided solid hints narrowing it to the dawn of computers, possibly with punchcards or subtle sounds
    – Laurent Duval
    Aug 25, 2016 at 5:31
  • You've certainly posted a legitimate answer, and it addresses the "why" question in interesting terms. I'm looking into the "when" question to see whether an answer to it offers any insight into the "why" question. If I turn up anything that seems useful, I'll post an answer.
    – Sven Yargs
    Aug 25, 2016 at 6:01

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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