Wilkins' Example, 1668

John Wilkins (1614 – 1672) was a clergyman, the first secretary of the Royal Society, and master of Trinity College, Cambridge, to which position he was appointed by his brother-in-law, Oliver Cromwell, at the urging of the Trinity College fellows. He had previously been the head of Wadham College, Oxford, where he nurtured a culture of scientific study that eventually led to the formation of the Royal Society.

In 1668 Wilkins published an exceedingly strange work outlining an attempt at a universal language—not “universal” like Esperanto, a language designed to be easily learned by speakers of a set of pre-existing languages, but rather a means of communication that would be free of ambiguity. The work is more closely related to present-day research in natural language processing by computers than to Roget's Thesaurus.

As part of his project, Wilkins addressed measure. Notice how uncannily close to the metric system Wilkins' units are, divided decimally from a fundamental unit of length of 39.25 inches (not far from 39.37).

By “longitude” Wilkins means the dimension length, or extension, not longitude as in “latitude and longitude.”

The metrological themes reflected in this selection are:

John Wilkins.
An Essay towards a real character, and a philosophical language.
London: Printed for Sa. Gellibrand, and for John Martyn, 1668.

The passage quoted is from Part II, chapter VIII, pages 191 and 192. The Scolar Press reprinted the book in facsimile in 1968. The entire book has been made available on the web both in facsimile, and typed in (a horrendous job, I speak from experience) at http://reliant.teknowledge.com/Wilkins/


II. Measures of Magnitude do comprehend both those of Length, and of Superficies or Area, together with those of Solidity; both comprehended in that which is adjoined, viz; the word CAPACITY, hold, contain. The several Nations of the World do not more differ in their Languages, then in the various kinds and proportions of these Measures. And it is not without great difficulty, that the Measures observed by all those different Nations who traffic together, are reduced to that which is commonly known and received by any one of them; which labor would be much abbreviated, if they were all of them fixed to any one certain Standard. To which purpose, it were most desirable to find out some natural Standard, or universal Measure, which hath been esteemed by Learned men as one of the desiderata in Philosophy. lf this could be done in Longitude, the other Measures might be easily fixed from thence.

This was heretofore aimed at and endeavored after in all those various Measures, derived from natural things, though none of them do sufficiently answer this end. As for that of a barley corn, which is made the common ground and original of the rest¹, the magnitude and weight of it may be so various in several times and places, as will render it incapable of serving for this purpose, which is true likewise of those other Measures, an Inch, Palm, Span, Cubit, Fathom, a Foot, Pace; &c. none of which can be determined to any sufficient certainty.

Some have conceived that this might be better done by subdividing a Degree upon the Earth: But there would be so much difficulty and uncertainty in this way as would render it unpracticable. Others have thought, it might be derived from the Quicksilver experiment² : But the unequal gravity and thickness of the Atmosphere, together with the various tempers of Air in several places and seasons, would expose that also to much uncertainty.

The most probable way for the effecting of this, is that which was first suggested by Doctor Christopher Wren, namely, by Vibration of a Pendulum: Time itself being a natural Measure, depending upon a revolution of the Heaven or the Earth, which is supposed to be everywhere equal and uniform. If any way could be found out to make Longitude commensurable to Time, this might be the foundation of a natural Standard. In order to which,

Let there be a solid Ball exactly round of some of the heaviest metals: Let there be a String to hang it upon, the smallest, limberest, and least subject to retch [sic. = stretch]: Let this Ball be suspended by this String, being extended to such a length, that the space of every Vibration may be equal to a second Minute of time, the String being, by frequent trials, either lengthened or shortened, till it attain to this equality: These Vibrations should be the smallest, that can last a sufficient space of time, to afford a considerable number of them, either 6, or 5003 at least; for which end, its passing an arch [sic. = arc] of five or six degrees at the first, may be sufficient. The Pendulum being so ordered as to have every one of its Vibrations equal to a second minute of time, which is to be adjusted with much care and exactness; then measure the length of this String, from its place of suspension to the Centre of the Ball; which Measure must be taken as it hangs free in its perpendicular posture, and not otherwise, because of stretching : which being done, there are given these two Lengths, viz. of the String, and of the Radius of the Ball, to which a third Proportional must be found out ; which must be as the length of the String from the point of Suspension to the Centre of the Ball is to the Radius of the Ball, so must the said Radius be to this third : which being so found, let two fifths of this third Proportional be set off from the Centre downwards, and that will give the Measure desired.4 And this (according to the discovery and observation of those two excellent persons: the Lord Viscount Brouncker, President of the Royal Society and Mon. Huygens, a worthy Member of it) will prove to be 38 Rhineland Inches, or (which is all one) 39 Inches and a quarter, according to our London Standard.

Let this Length therefore be called the Standard; let one Tenth of it be called a Foot : one Tenth of a Foot, an Inch; one Tenth of an Inch, a Line. And so upward, Ten Standards should be a Perch ; Ten Perches, a Furlong; Ten Furlongs, a Mile; Ten Miles, a League, &c.

And so for Measures of Capacity: The cubical content of this Standard may be called the Bushel: the Tenth part of the Bushel, the Peck; the Tenth part of a Peck, a Quart; and the Tenth of that, a Pint, &c. And so for as many other Measures upwards as shall be thought expedient for use.

As for Measures of Weight; Let this cubical content of distilled Rainwater be the Hundred, the Tenth Part of that, a Stone; the Tenth part of a Stone, a Pound; the Tenth of a Pound, an Ounce; the Tenth of an Ounce, a Dram; the Tenth of a Dram, a Scruple; the Tenth of a Scruple, a Grain, &c. And so upwards; Ten of these cubical Measures may be called a Thousand, and Ten of these Thousand may be called a Tun, &c.

As for the Measures of Money, 'tis requisite that they should be determined by the different Quantities of those two natural Metals which are the most usual materials of it, viz. Gold and Silver, considered in their Purity without any alloy. A Cube of this Standard of either of these Metals may be styled a Thousand or a Talent of each; the Tenth part of this weight, a Hundred; the Tenth of a Hundred, a Pound; the Tenth of a Pound, an Angel; the Tenth of an Angel, a Shilling; the Tenth of a Shilling, a Penny; the Tenth of a Penny, a Farthing.

I mention these particulars, not out of any hope or expectation that the World will ever make use of them, but only to show the possibility of reducing all Measures to one determined certainty.


1. The English units of length have been derived legally by subdivision of the yard as far back as records take us. If anyone really ever lined up barleycorns to make an inch, it would have been country folk, and it is hard to see why they would do this. In England, the origin of the unit of mass called the “grain” is traceable to wheat rather than barley.

2. Quicksilver experiment. Apparently this refers to using the height of the mercury in a barometer as a standard of length.

3. That is, “600, or 500 at the least.” This usage sounds odd today, while in “five or six hundred at the most” we instantly recognize that the five is five hundred.

4. We have been unable to discover the source of, or reason for, Wilkins' method. Provided the pendulum's swing is kept to four or five degrees, the Galilean relationship between length and period applies and (the mass of the string being negligible), the length is the distance from the point of suspension to the center of the spherical bob. A search for publications by Huygens or Brouncker that contain the procedure Wilkins' describes has been fruitless. Since the correction depends on the radius of the bob, is it intended to compensate for air resistance? If anyone recognizes his procedure, please email us.

Wilkins' correction is very small: assuming a bob with a 3-inch radius, and 36 inches from the center of the bob to the point of suspension, the correction would be 0.1 inches. The length of the seconds pendulum in Greenwich is actually 31.1398 inches, about 0.1 inch less than the 39.25 inches Wilkins gives.

home | library index | search |  contact drawing of envelope |  contributors | 
help | privacy | terms of use