When looking through the National Physical Laboratory's SI unit conventions, I have found a strange clause in their list:

  • For unit values more than 1 or less than -1 the plural of the unit is used and a singular unit is used for values between 1 and -1.

If you follow this rule through, you would condemn the following usage:

UTC is kept always within 0.9 seconds of UT1 by the insertion of extra seconds as necessary (known as positive leap seconds). It could happen that seconds would need to be removed (negative leap seconds), but so far all leap seconds have been positive.

This quote has come directly from an article by the NPL about leap seconds showing that even they don't follow their own rules.
Google Ngram Viewer contains several false positives but places "0.1 second" above "0.1 seconds" but "0.2 second(s)" is the other way round.

Is this rule kept to?
Does it have any bearing in scientific English?

  • 4
    Just to confuse matters a little more, some people would say 0.9 of a second! Jones was 0.9 of second faster than Smith, in the 400 metres. – WS2 Dec 30 '16 at 14:13
  • 3
    The problem disappears if seconds in that quote is not referring to the SI unit -- perhaps it's referring to a CGS second. Certainly the usual usage is for the singular noun to be used only for exactly one of them. – Andrew Leach Dec 30 '16 at 14:22
  • 1
    @AndrewLeach Yes, naturally I would only use the grammatical singular for -1 or 1 unit. However, I would say 1.0 (one point zero) units. – BladorthinTheGrey Dec 30 '16 at 14:26
  • @WS2 Thinking about it, I would usually only say of a second if I omitted the zero before the point. I.e. Jones was point nine of a second rather than Jones was zero point nine of a second. Would you also follow that rule? – BladorthinTheGrey Dec 30 '16 at 14:29
  • 2

At the computer magazines where I worked for many years, we had a strict house rule to identify measurements of less than one with a singular unit of measure. For example, we would say

¾ inch or 0.75 inch


¾ inches or 0.75 inches

This style decision required endless enforcement because many of our writers habitually used plural units of measure with measurements of less than one.

One style guide that shares our old house style preference is Words into Type, third edition (1974), which offers this advice:

Decimals. For quantities of less than one, a zero should be set before the decimal point except for quantities that never exceed one. Also, when units are not abbreviated the singular is used for quantities of one or less. ... [Relevant examples:] 0.32 second, 0.75 grain

The Chicago Manual of Style, sixteenth edition (2010) evades this question almost completely. But to the extent that it has anything useful to say on the subject, it tends to contradict Words into Type. Here is the most direct remark it makes on the subject:

10.68 Plural form for abbreviations of US measure. Abbreviations of US units of measure, like their scientific counterpart, are identical in the singular and plural. ... Note that the unit of measure in such expressions as 0.5 yd. and 1.5 yd. is generally pronounced as if it were plural (i.e., point five yards; one point five yards).

This last statement seems to suggest that Chicago endorses writing out 0.5 yd. as 0.5 yards rather than as 0.5 yard—although it never explicitly says so. It's a very odd treatment, when you think about it: Chicago explains how 0.5 yd. "is generally pronounced" by US English speakers but never specifies whether US writers should spell out 0.5 yd. as 0.5 yard or as 0.5 yards. From this episode, you might think that the book was called The Chicago Manual of Pronunciation Style.

The Oxford Guide to Style (2002) sides with Words into Type on this matter:

7.2 Units ... Elsewhere [that is, aside from instances where units help form compound adjectives, as in "a two-mile walk"], units are pluralized as necessary, but not if the quantity or number is less than one: two kilos, three miles, 0.568 litre, half a pint.

If I recall correctly, part of the rationale for using singular units of measure for quantities less than one was that it helped avoid misreadings of, for example, ".8 pound[s]" as "8 pounds." (That is also a sound reason for including the zero before the decimal point, by the way.) But more than anything else, I think, we were swayed by the idea that a number consisting of a fraction less than one would certainly be attached to a singular unit (for example, ⅝ inch), so it wouldn't make sense to treat the exact decimal equivalent of ⅝ (0.625) as taking a plural unit of measure (inches) instead of a singular one (inch). But obviously, reasonable (and unreasonable) people differ on this question.

  • 2
    +1 As you are no doubt aware, this is most critical with on-board intercoms and wireless transmissions, and the wrong usage can result in critical error. The telling point in your argument for me is "a sound reason for including the zero before the decimal point"--as a former USN EM (nukie) and later BSEET, it somehow still feels odd to say "point eight inch": My gut tends to go with inches. So saying "zero point eight inches" (emphasis 0) is more natural IMHO, and avoids any error. And when you are dealing with nukes, or a coordinate for dropping a mortar round, that can be kinda important. – Cascabel Jan 3 '17 at 5:30

I think the rule as written differs significantly from the common usage.

It is normal to use the plural "seconds" in all cases.

As @WS2 identified, the phrase "of a second" is also fairly common when refering to a fraction of a second less than 1.

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.