# Central dot as decimal point in top journal

The Lancet, one of the top medical journals in the world, requires 'Type decimal points midline (ie, 23·4, not 23.4)'. Does anyone know why this is? As a mathematician, I'll always read 23·4 as 92. Seems odd to me that such a high profile paper would insist on such a strange formatting choice, which goes against the SI standard, so I'm guessing there's a good reason for it.

• Not really an answer, but there is precedent for use of a midline dot as a decimal point: en.wikipedia.org/wiki/Interpunct Oct 5 '18 at 15:30
• When some of us hand write numbers with decimals (gasp! Pens & pencils still exist) we were taught to put the decimal point in the middle, as it could get “lost” when using lined paper... Oct 5 '18 at 16:20
• I have only one paper in a British journal (Proceedings of the London Mathematical Society). Their copy-editor changed all the decimals in the paper. If I used 5.243, that was changed to 5･243 . Oct 5 '18 at 19:38
• "...which goes against the SI standard..." There are many things in the world going against the SI standard. Oct 6 '18 at 19:25
• When I was in school (in the U.K.) we wrote the dot in the middle of the line for decimals. When we got to dot products we wrote them with a lower dot. When I went to university we wrote a middle dot for dot products and multiplication and I’m not sure I ever wrote a number with a decimal point but I guess I’d use a lower dot. Oct 7 '18 at 10:33

This notation was more common historically, particularly in the British empire. My guess would be that the Lancet, being an old journal founded in England in 1823, is sticking with it because of tradition rather than a really good reason.

• It would seem to me that tradition is a good reason.
– MaxW
Oct 7 '18 at 17:00
• @MaxW It's not a terrible reason on its own, and I'm not calling for an end to all traditions here, but if there are good reasons to stop or continue a practice, well, those reasons generally trump "we've always done it that way". Note that, for the purposes of this answer, I accept OP's premise that the notation is unusual and somewhat confusing. Oct 7 '18 at 20:42
• Yep, I have seen this in other traditional British journals as well (London Soc or Edinburgh Math Soc maybe). Jul 5 '19 at 11:39
• @darijgrinberg Do they still use it? Proc. R. Soc. A and Phil. Mag. seem to have moved away from it, for example. Jul 6 '19 at 15:54
• Actually, it seems that they are phasing it out. This paper, for example, has bottom-aligned dots for decimal numbers (see page 3) but still has centered $\cdot$s for "Theorem 1$\cdot$2" etc. Jul 7 '19 at 5:07

To expand on @Anyon's answer more generally, mathematical notation is not universal between countries. Where the decimal symbol appears—and what the decimal symbol even is—varies significantly. For instance, in most of continental Europe, the period and comma are switched so that what in the US would be "24,321.12" would be "24.321,12" in Germany or the Netherlands. Similarly, the "dot product" is sometimes written on the baseline rather than on the center line.

So this is just an expression of a stylistic preference or tradition.

• Sometimes the dot product is even written with a cross (and the cross product with a wedge). This confused me a hell of a lot in high school Oct 5 '18 at 16:17
• In fact, modern scientific guides suggest to use a small space instead of a comma or a dot to separate thousands. Oct 5 '18 at 18:34
• @MassimoOrtolano This is the SI standard since 2003 to not use the comma or dot for anything but the decimal separator (bipm.org/en/CGPM/db/22/10)
– LCT
Oct 5 '18 at 18:45
• That rather stuffs up the CSV file then... Oct 5 '18 at 19:01
• @JAB to add to the variation, in Russian literature dot product is commonly written as (a,b) (example), while cross product is [a,b] (example). And this despite comma being decimal separator. What a mess. Oct 7 '18 at 9:40

The unspaced centred dot, as in 23·4 (= 234 ÷ 10), means the decimal point. To denote multiplication, spacing is necessary: 23 · 4 = 92. As long as one sticks to this convention, no confusion arises. From some people's point of view, it is nice to distinguish thus the marker for decimal point from those for multiplication and the end of a sentence.

• This sounds like a very error-prone convention. Oct 8 '18 at 6:15
• I didn't downvote BTW. It's good to know this convention exists, even if it's a dangerous one. Oct 8 '18 at 9:57
• -1'd because I disagree that this avoids confusion, though this answer does seem otherwise helpful in describing the thinking. It does seem a tad strange to have such a convention based on spacing when spacing already distinguishes decimal points from periods; and while someone might well argue that the existing convention could lead to confusion in some cases, it's strange to then invent a new convention that has the exact same weakness.
– Nat
Mar 1 '19 at 8:35

All mathematical text-books published in the UK used the interpunct for the decimal indicator. It was not only pupils in school! (I am old enough to remember pupils being reprimanded for failing to place the decimal point accurately.) I wonder why and exactly when we made the change to the decimal point on the line. Was it in the 1980s, when people started to use computer keyboards? Specialist knowledge is required (and a numeric keypad) to produce an interpunct · and this could well be the explanation. This change is at least something for which we cannot blame the EU!

• And even though it is possible to produce an interpunct via keycode, numeric input to most software must use the ordinary key. May 18 '19 at 23:04