# Fahrenheit - Cecil missed something **URL:** \ **Category:** Cecil's Columns/Staff Reports\ **Created:** [January 24, 2008, 6:28pm UTC](https://boards.straightdope.com/t/fahrenheit-cecil-missed-something/434839 "2008-01-24T18:28:08Z")\ **Posts on this page:** 20\ **Page:** 2
**Author:** ![DSYoungEsq](https://avatars.discourse-cdn.com/v4/letter/d/c6cbf5/32.png) [@DSYoungEsq](https://boards.straightdope.com/u/DSYoungEsq)\ **Post date:** [February 4, 2008, 5:46am UTC](https://boards.straightdope.com/t/fahrenheit-cecil-missed-something/434839/21 "2008-02-04T05:46:10Z")
[QUOTE=foolsguinea] :dubious: Really. Which is more credible?[ul][li]That the true & perfect statistical average is within a 20th of a degree Fahrenheit, or[\*]that given a typical variation of over 3 degrees Celsius (over 5 degrees Fahrenheit), Mr Celsius used a whole number in the middle of the range & considered further significant digits meaningless?[/ul]Good grief! What would Ockham say?OK, so you have finely tuned temperature sense. You could then sense a shift of half a degree Celsius. That doesn’t change the fact that human body temperature varies more than a degree within a single body, & ambient temperature varies even more widely than that within a room, & even more from one side of a street to another. Your thermometer is showing a precise temperature where it is, which has a deviation of a couple of degrees from where you are. It’s interesting detail, but it’s sort of excessive.[/li] And I like Fahrenheit, with its arbitrariness & non-decimal significant points, & sort of feel the “100° = body temperature” has a weird animal elegance that basing a scale on the boiling temperature of water lacks. But I know this is mere cultural prejudice, & I think people protest too much on its behalf. [/QUOTE] You totally miss the point about accuracy. 98.6 is EXACTLY equal to 37. The accuracy of 37 is what is in issue, but when we accept that 37 is the statistical average, then it is equally accurate to say the 98.6 is the same statistical average. It would be highly improper to assert that 99 was the same statistical average, because it would NOT be. Now, if the measurements upon which the assertion was based had been done with the Farenheit scale, we might well assert something like 98 or 99 as the “average human body temperature taken orally.” But it wasn’t. So get over it. 😛 ---
**Author:** ![Chronos](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/chronos/32/134_2.png) [@Chronos](https://boards.straightdope.com/u/Chronos)\ **Post date:** [February 4, 2008, 6:20am UTC](https://boards.straightdope.com/t/fahrenheit-cecil-missed-something/434839/22 "2008-02-04T06:20:57Z")
Here’s the problem: 98.6 may be accurate, but it’s not precise. Or at least, it’s not as precise as the number of digits implies. If I say that a particular temperature is 98.6 degrees, then I mean that it’s somewhere between 98.55 and 98.65. If, on the other hand, I say that a particular temperature is 37 degrees, then I mean that it’s somewhere between 36.5 and 37.5. The mistake here is similar to that of the museum janitor, who says that a particular fossil is 65,000,003 years old, because he heard it was 65 million years old when he started working there, and he’s been there for three years now. ---
**Author:** ![John\_W.Kennedy](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/john_w.kennedy/32/1031_2.png) [@John\_W.Kennedy](https://boards.straightdope.com/u/John_W.Kennedy)\ **Post date:** [February 4, 2008, 6:28pm UTC](https://boards.straightdope.com/t/fahrenheit-cecil-missed-something/434839/23 "2008-02-04T18:28:53Z")
Strictly speaking, it’s not “not precise”, but _too_ precise – that is, it’s more precise than is the reality it claims to represent. To make this a bit easier to explain, though, could we get a sigma or so to use in giving examples? ---
**Author:** ![Aimew](https://avatars.discourse-cdn.com/v4/letter/a/91b2a8/32.png) [@Aimew](https://boards.straightdope.com/u/Aimew)\ **Post date:** [September 7, 2012, 7:09pm UTC](https://boards.straightdope.com/t/fahrenheit-cecil-missed-something/434839/24 "2012-09-07T19:09:34Z")
> [@GreenWyvern](#): > > I thought that the significance of the number 180 was so immediately obvious that it didn’t need any further comment. Obviously, I was wrong. Ever use a protractor? Ever hear the expression “180-degree turn” ? > > The numbers 60, 180, 360 have been used for divisions of time and space for thousands of years. > > You noticed that 180 is divisible by 10. If you wok it out, you’ll find that it’s actually divisible by 2, 3, 4, 5, 6, 9, 10, 12, 15, 18, 20, 30, 36, 45, 90. That’s why it’s useful, especially if you don’t have a calculator. > > My point is that the article is _wrong_. > > Fahrenheit based his scale on the freezing point of water and the boiling point of water, nothing else. He set the difference as 180 - an obvious number to anyone familiar with geometry or astonomy. He then made the starting point 32 degrees further down in order to avoid negative numbers, which were still quite esoteric to ordinary people at that time. This is the story I have heard for most of my 63 years. I think I first found it in one of Asimov’s non-fiction books on physics when I was a young man, in my twenties, I think. I was surprised that Cicil didn’t seem to know that one; he is usually quite accurate. Oh well, nobody is perfect; but he comes closer than most. ---
**Author:** ![Exapno\_Mapcase](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/exapno_mapcase/32/1051_2.png) [@Exapno\_Mapcase](https://boards.straightdope.com/u/Exapno_Mapcase)\ **Post date:** [September 7, 2012, 7:29pm UTC](https://boards.straightdope.com/t/fahrenheit-cecil-missed-something/434839/25 "2012-09-07T19:29:16Z")
> [@Aimew](#): > > This is the story I have heard for most of my 63 years. > I think I first found it in one of Asimov’s non-fiction books on physics when I was a young man, in my twenties, I think. > I was surprised that Cicil didn’t seem to know that one; he is usually quite accurate. Oh well, nobody is perfect; but he comes closer than most. To my surprise, Asimov honestly did say this. It’s in his essay, The Height of Up," from Oct. 1959 F&SF and reprinted in View from A Height and Asimov on Physics. I don’t even have to type it all out, because David Simmons did it in [this 2004 thread](http://www.boards.straightdope.com/sdmb/showthread.php?t=263356): > [@Degree as a unit of temperature?](https://boards.straightdope.com/t/degree-as-a-unit-of-temperature/252101/17): > > Here is Isaac Asimov’s take on the 180[sup]o[/sup] difference and the Fahrenheit temperature scale in general. It’s from his essay _The Height of Up._ By this account Fahrenheit originally set his high mark at body temperature of 12 but changed it to 96 because his thermometer was capable of greater accuracy than had been the case until then. On this scale freezing water was a little under 32 and boiling water was a little under 212, with a difference of not quite 180[sup]o[/sup]. Fahrenheit possibly liked 180 exactly and since it was so close that that anyway, he changed his reference points to two physical phenomena of water. He set the freezing point at exactly 32 and the boiling point exactly 180 away at 212. This made body temperature 98.6, but that’s not a constant and only approximate anyone so it doesn’t matter. > > I can’t find any reference to Fhrenheit setting body temperature at 100. > > \*"But then, in 1714, a German physicist named Gabriel Daniel Fahrenheit made a major step forward. The liquid that had been used in the early thermometers was either water or alcohol. Water, however, froze and became useless at temperatures that were not very cold, while alcohol boiled and became useless at temperatures that were not very hot. What Fahrenheit did was to substitute mercury. Mercury stayed liquid well below the freezing point of water and well above the boiling point of alcohol. Furthermore, mercury expanded and contracted more uniformly with temperature than did either water or alcohol. Using mercury, Fahrenheit constructed the best thermometers the world had yet seen. > > With his mercury thermometer, Fahrenheit was now ready to use Newton’s suggestion[sup]\*[/sup]; but in doing so, he made a number of modifications. He didn’t use the freezing point of water for his zero (perhaps because winter temperatures below that point were common enough in Germany and Fahrenheit wanted to avoid the complication of negative temperatures). Instead, he set zero at the very lowest temperature he could get in his laboratory, and that he attained by mixing salt and melting ice. Then he set human body temperature at 12, following Newton, but that didn’t last either. Fahrenheit’s thermometer was so good that a division into twelve degrees was unnecessarily coarse. Fahrenheit could do eight times as well, so he set body temperature at 96. > > On this scale, the freezing point of water stood at a little under 32, and the boiling point at a little under 212. It must have struck him as fortunate that the difference between the two should be about 180 degrees, since 180 was a number that could be divided evenly by a large variety of integers including 2, 3, 4, 5, 6, 9, 10, 12, 15, 18, 20, 30, 36, 45, 60 and 90. Therefore, keeping the zero point as was, Fahrenheit set the freezing point of water at exactly 32 and the boiling point at exactly 212. That made body temperature come out (on the average) at 98.6°, which was an uneven value, but this was a minor point. > Thus was born the Fahrenheit scale, which we, in the United States, use for ordinary purposes to this day. We speak of “degrees Fahrenheit” and symbolize it as “O F.” so that the normal body temperature is written 98.6° F."\* > > - In 1701 Isaac Newton suggested that the temperature scale start with the temperature of melting ice as the zero point and the body temperature at 12. Well, that should take Asimov’s reputation down a peg. ---
**Author:** ![Aimew](https://avatars.discourse-cdn.com/v4/letter/a/91b2a8/32.png) [@Aimew](https://boards.straightdope.com/u/Aimew)\ **Post date:** [September 7, 2012, 7:29pm UTC](https://boards.straightdope.com/t/fahrenheit-cecil-missed-something/434839/26 "2012-09-07T19:29:26Z")
> [@DSYoungEsq](#): > > Well, they weren’t arrived at by worrying one jot about the number 180. First of all, just because it’s a cool number is meaningless. Unlike some things (hours of a day, e.g.), it isn’t important to be able to divide a temperature difference up into parts like that. What purpose would it serve? Second of all, we have no evidence that 180 was in their thinking. We DO have evidence that the adjustment was made simply to allow for an “easy” conversion factor, given that 32° F was going to stay as the freezing point of water. In short, they looked at the scale as calibrated by Farenheit, which had no easy conversion factor, and they said, “Hey, you know what? If we set the boiling point at 212°, we can use 9:5 as a conversion factor!” The only part 180 played in this was the fortuitous happenstance of being 9/5ths of 100. > > I’m glad you dropped by; we always enjoy new vict…, er, that is, new guests. 🙂 However, if you decide to stay, you will quickly find that what you did here, which is simply speculate on an answer, based on what seemed to be a good idea to you, simply isn’t an acceptable method of proof here. We like to rely on things like evidence (when we aren’t relying on our own big egos :smack: ). 😉 Fahrenheit was looking for a scale for his thermometer for meteorological purposes. The 180, while being a convenient number in and of itself, also allowed for fine measurements for the outside temperature. He (supposedly) added 32 to avoid (most) negative numbers as lay people didn’t understand them readily. ---
**Author:** ![TriPolar](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/tripolar/32/3008_2.png) [@TriPolar](https://boards.straightdope.com/u/TriPolar)\ **Post date:** [September 7, 2012, 7:46pm UTC](https://boards.straightdope.com/t/fahrenheit-cecil-missed-something/434839/27 "2012-09-07T19:46:25Z")
> [@Aimew](#): > > Fahrenheit was looking for a scale for his thermometer for meteorological purposes. The 180, while being a convenient number in and of itself, also allowed for fine measurements for the outside temperature. He (supposedly) added 32 to avoid (most) negative numbers as lay people didn’t understand them readily. Yes. Fahrenheit is very useful for the temperature range found in most of the temperate zone. I recall he did make adjustments to get the 180, and other points. For instance his original 100F was human body temperature. But that left a fractional number for the boiling and freezing points. It was easier to set those as integers and leave body temperature fractional (and body temperature isn’t as consistent as boiling and freezing anyway). ---
**Author:** ![John\_W.Kennedy](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/john_w.kennedy/32/1031_2.png) [@John\_W.Kennedy](https://boards.straightdope.com/u/John_W.Kennedy)\ **Post date:** [September 7, 2012, 11:39pm UTC](https://boards.straightdope.com/t/fahrenheit-cecil-missed-something/434839/28 "2012-09-07T23:39:48Z")
Asimov was a biochemist, not a historian; I dare say he relied on secondary sources. ---
**Author:** ![Exapno\_Mapcase](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/exapno_mapcase/32/1051_2.png) [@Exapno\_Mapcase](https://boards.straightdope.com/u/Exapno_Mapcase)\ **Post date:** [September 8, 2012, 2:26pm UTC](https://boards.straightdope.com/t/fahrenheit-cecil-missed-something/434839/29 "2012-09-08T14:26:52Z")
> [@John\_W.Kennedy](#): > > Asimov was a biochemist, not a historian; I dare say he relied on secondary sources. That’s not the issue here. Of course he relied on secondary sources, but he should damn well have understood the point made by RedSwinglineOne, that the divisibility of 180 is meaningless when applied to temperature on a non-absolute zero scale. What probably happened is that he was thinking about five other essays he was working on that did make a point about the divisibility of 180 and why 360 was a favorite number of the ancients and just automatically copied it into a place where it was more than irrelevant. ---
**Author:** ![John\_W.Kennedy](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/john_w.kennedy/32/1031_2.png) [@John\_W.Kennedy](https://boards.straightdope.com/u/John_W.Kennedy)\ **Post date:** [September 8, 2012, 9:39pm UTC](https://boards.straightdope.com/t/fahrenheit-cecil-missed-something/434839/30 "2012-09-08T21:39:51Z")
As has already been pointed out, multiplication and division comes into the issue of making thermometers. (Don’t try to catch me on absolute zero; I caught my 7th-grade science teacher on that one over half a century ago.) ---
**Author:** ![Exapno\_Mapcase](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/exapno_mapcase/32/1051_2.png) [@Exapno\_Mapcase](https://boards.straightdope.com/u/Exapno_Mapcase)\ **Post date:** [September 8, 2012, 9:47pm UTC](https://boards.straightdope.com/t/fahrenheit-cecil-missed-something/434839/31 "2012-09-08T21:47:38Z")
> [@John\_W.Kennedy](#): > > As has already been pointed out, multiplication and division comes into the issue of making thermometers. (Don’t try to catch me on absolute zero; I caught my 7th-grade science teacher on that one over half a century ago.) I don’t see it. What are you you referring to? ---
**Author:** ![gamerunknown](https://avatars.discourse-cdn.com/v4/letter/g/6bbea6/32.png) [@gamerunknown](https://boards.straightdope.com/u/gamerunknown)\ **Post date:** [September 9, 2012, 9:50am UTC](https://boards.straightdope.com/t/fahrenheit-cecil-missed-something/434839/32 "2012-09-09T09:50:40Z")
> [@Exapno Mapcase](#): > > What are you you referring to? I believe it was this post. > [@russl5445](#): > > You take your brand new (unmarked) thermometer and measure human body temperature, and the temperature where water freezes. Make little marks on the thermometer for each. It’s easy to then find the midpoint between the two marks - that’s 64 degrees. You can then measure out exactly where 0 degrees should be. You then divide the regions in half, over and over. Since 32 and 64 are both powers of two, it’s easy and accurate. Seems like a “just so” story in the absence of any cites, especially due to the variability of human body temperature. If markings for each temperature were tied to the measure of body heat, then thermometer readings would vary quite widely. Unless there were some form of measurement concordance, where they marked where some other thermometer gave a reading of 96 degrees as 96 degrees on each new one. ---
**Author:** ![Exapno\_Mapcase](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/exapno_mapcase/32/1051_2.png) [@Exapno\_Mapcase](https://boards.straightdope.com/u/Exapno_Mapcase)\ **Post date:** [September 9, 2012, 5:09pm UTC](https://boards.straightdope.com/t/fahrenheit-cecil-missed-something/434839/33 "2012-09-09T17:09:55Z")
> [@gamerunknown](#): > > I believe it was this post. > > Seems like a “just so” story in the absence of any cites, especially due to the variability of human body temperature. If markings for each temperature were tied to the measure of body heat, then thermometer readings would vary quite widely. Unless there were some form of measurement concordance, where they marked where some other thermometer gave a reading of 96 degrees as 96 degrees on each new one. That doesn’t really apply to 180. You’d think 256 would be far more useful in that context. Or 128 for body temperature. And it’s not what Fahrenheit or anybody did. ---
**Author:** ![TriPolar](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/tripolar/32/3008_2.png) [@TriPolar](https://boards.straightdope.com/u/TriPolar)\ **Post date:** [September 9, 2012, 5:24pm UTC](https://boards.straightdope.com/t/fahrenheit-cecil-missed-something/434839/34 "2012-09-09T17:24:12Z")
Fahrenheit designed a horse and ended up with a camel. He made adjustments to a concept to end up with something that had integer representations for the freezing and boiling points of water, under typical conditions. A lot of other niceties came along with the scale like the typical range of temperatures in a lot of the world on a scale of 0 to 100, numbers like 180 that some find handy, typical body temperature near 100. Fahrenheit may or may not have considered any of these factors, but historians should be accurate in reporting the known methodology for arriving at the scale though. But without these docudrama revisions of history Cecil wouldn’t have many column topics. ---
**Author:** ![Kenm](https://avatars.discourse-cdn.com/v4/letter/k/bc79bd/32.png) [@Kenm](https://boards.straightdope.com/u/Kenm)\ **Post date:** [September 9, 2012, 7:29pm UTC](https://boards.straightdope.com/t/fahrenheit-cecil-missed-something/434839/35 "2012-09-09T19:29:29Z")
The Celsius scale has its shorcomings, not the least of which suggest that Anders Celsius was a few degrees short of body-normal temperature: He made the boiling point of water 0 and the freezing point 100. Also, Wikipedia: > [@](#): > > What is often confusing about the Celsius measurement is that it follows an interval system but not a ratio system; that it follows a relative scale not an absolute scale. This is put simply by illustrating that while 10 °C and 20 °C have the same interval difference as 20 °C and 30 °C the temperature 20 °C is not twice the air heat energy as 10 °C. Question: Could Asimov’s reputation be saved if the 180° business, along with its many evenly divided numbers, mean that through their use in developing the scale Fahrenheit avoided Celsius’s absolute-temperature failure? BTW, The Wiki article needs an editor who could ask the author questions directly. It suffers from some strange syntax here and there, as if English is not the author’s first language. If so, that may have introduced errors inadvertantly. ---
**Author:** ![Powers](https://avatars.discourse-cdn.com/v4/letter/p/edb3f5/32.png) [@Powers](https://boards.straightdope.com/u/Powers)\ **Post date:** [September 11, 2012, 1:46am UTC](https://boards.straightdope.com/t/fahrenheit-cecil-missed-something/434839/36 "2012-09-11T01:46:26Z")
> [@Kenm](#): > > Question: Could Asimov’s reputation be saved if the 180° business, along with its many evenly divided numbers, mean that through their use in developing the scale Fahrenheit avoided Celsius’s absolute-temperature failure? No; the only absolute temperature scale (that is, one in which a doubling of the temperature value represents a doubling in heat energy) in wide use is Kelvin, though it is at least a centigrade scale like Celsius is. Powers &8^] ---
**Author:** ![John\_W.Kennedy](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/john_w.kennedy/32/1031_2.png) [@John\_W.Kennedy](https://boards.straightdope.com/u/John_W.Kennedy)\ **Post date:** [September 11, 2012, 1:59am UTC](https://boards.straightdope.com/t/fahrenheit-cecil-missed-something/434839/37 "2012-09-11T01:59:05Z")
> [@Exapno\_Mapcase](#): > > That doesn’t really apply to 180. You’d think 256 would be far more useful in that context. Or 128 for body temperature. And it’s not what Fahrenheit or anybody did. Several thousand years of human history rather suggests that people prefer 180 to 128 (unless they’re computer professionals). ---
**Author:** ![Irishman](https://avatars.discourse-cdn.com/v4/letter/i/b487fb/32.png) [@Irishman](https://boards.straightdope.com/u/Irishman)\ **Post date:** [September 11, 2012, 9:45pm UTC](https://boards.straightdope.com/t/fahrenheit-cecil-missed-something/434839/38 "2012-09-11T21:45:48Z")
> [@DSYoungEsq](#): > > the relevant text of which can be found [here](http://www.sizes.com/units/temperature_Fahrenheit.htm). For clarity, it took me a moment to realize the relevant text is on the tab labeled “Sources”. For some reason when I go to a link, I expect to be taken to the correct tab. 😉 > [@DSYoungEsq](#): > > You totally miss the point about accuracy. > > 98.6 is EXACTLY equal to 37. The accuracy of 37 is what is in issue, but when we accept that 37 is the statistical average, then it is equally accurate to say the 98.6 is the same statistical average. It would be highly improper to assert that 99 was the same statistical average, because it would NOT be. This entirely misses the point. “37” is the average to the ones place, but that says nothing about what numbers were actually used. Suppose Celsius took measurements ranging from 35.7 to 39.0 deg for human body, and then calculated an statistical average of 37.2 deg. However, he noted that the range was so broad that worrying about that 0.2 was rather pointless, so he conveniently dropped it and said “average human temp = 37”. If you take his actual calculated value to the same decimal precision as we use in Fahrenheit, then you have to start with the 37.2 deg. (37.2 \*9 / 5) + 32 = 98.96, which rounds for 3 decimal places to 99.0 deg F. “37” could be any value from 36.5 to 37.4 and round to 37 when looking at 2 significant digits (or even degrees of Celsius). We have no information on the precision of Celsius’s data for determining 37, so we cannot say if his numerical average was 37.000, or if it was 37.2 and he just decided 37 was more convenient. > [@Powers](#): > > No; the only absolute temperature scale (that is, one in which a doubling of the temperature value represents a doubling in heat energy) in wide use is Kelvin, though it is at least a centigrade scale like Celsius is. Centigrade meaning "the temperatures between water freezing and boiling points is divided into 100 equal increments). Yes, Kelvin scale uses Celsius degree divisions but moves the zero point to absolute zero. The Fahrenheit equivalent is the Rankine scale - it uses Fahrenheit degrees but puts zero at absolute zero. In practice it probably isn’t very widely used, other than for college student homework assignments. ---
**Author:** ![Exapno\_Mapcase](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/exapno_mapcase/32/1051_2.png) [@Exapno\_Mapcase](https://boards.straightdope.com/u/Exapno_Mapcase)\ **Post date:** [September 12, 2012, 1:53am UTC](https://boards.straightdope.com/t/fahrenheit-cecil-missed-something/434839/39 "2012-09-12T01:53:49Z")
> [@John\_W.Kennedy](#): > > Several thousand years of human history rather suggests that people prefer 180 to 128 (unless they’re computer professionals). Not for temperature. No one in history pre-Fahrenheit did. Nor did he, really. ---
**Author:** ![John\_W.Kennedy](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/john_w.kennedy/32/1031_2.png) [@John\_W.Kennedy](https://boards.straightdope.com/u/John_W.Kennedy)\ **Post date:** [September 13, 2012, 1:14am UTC](https://boards.straightdope.com/t/fahrenheit-cecil-missed-something/434839/40 "2012-09-13T01:14:57Z")
> [@Irishman](#): > > The Fahrenheit equivalent is the Rankine scale - it uses Fahrenheit degrees but puts zero at absolute zero. In practice it probably isn’t very widely used, other than for college student homework assignments. It’s in the CRC book, so I’m sure it was used somewhere once. So, I dare say, was the Réaumur scale (0˚= freezing, 80˚= boiling). [Previous page](https://boards.straightdope.com/t/fahrenheit-cecil-missed-something/434839.md?page=1) [Next page](https://boards.straightdope.com/t/fahrenheit-cecil-missed-something/434839.md?page=3)