The arrangement may make it convenient for Americans to venture abroad, but it also contributes to cultural isolation. And the isolation extends to the system of measurement.

There have been an undue number of letters to the editor of this magazine over the years criticizing the metric system for its inconsistency (as compared to the English system, presumably).

We little realize how inconsistent, illogical, and unsystematic our own "system" is, because we are not readily confronted with alternatives.

The inconsistencies that critics of the metric system focus on are minuscule and most are due to the system's evolution. The evolution is inevitable. Any system that wants to keep up with an evolving society must change.

Among metric users, as in any population, there are people who do not want to accept change, or are not aware of revisions or the need for them, who prefer to cling to established ways. Inconsistencies arise, then, in the way that people use the system, not in the system itself.

There indeed is one, and only one, metric system—always the latest revision of the international standard that describes it. For engineers worldwide, the standard is ISO 1000. This standard is subject to change, as any standard must be, and a user is expected to follow the latest version, which is the practice with all standards.

The standard describes a universal, international language of measurement.

Essentially, all units created in modern times are metric in every country of the world, including the United States. The evolution is coordinated by an international committee in which the United States has participated since 1875.

It may be comforting to many to learn that the standard, in the section most people use, is not expected to change for a long time. And the last major revision took place in 1960—two generations ago.

Note that there is no officially recognized forum for the development of any other system of units, including the English system or any version of it.

The modern system of measurement is properly called SI, not metric. SI, for the French Système International, is built on seven arbitrary units, called base units. Individually, they measure such basic physical quantities as length, mass, or time. Alone or in combination, they let man-kind measure anything.

Derived Units

There are, of course, hundreds of units needed for measuring "anything," but they are all derived from those seven. The derivation is done in a way that provides the marvelous and unique feature of SI: There are no conversion factors.

Some of the derivations were given a special name. This was done in cases where the combination would be too long and cumbersome for frequent use, or where confusion could result. Most have been used in the English system for generations because no official non-metric equivalent ever existed.

All derived units can be expressed in terms of the base units. For example, the SI unit for force, called the newton, is derived from mass times acceleration, the kilogram accelerating a meter per second per second or, in graphic symbols, kg•m/s².

Most derived units are as straightforward as the relationship of length to breadth to compute area, m2. Knowing that torque means force times distance leads to the newton meter, N•m. Or, for energy density (energy per mass or volume), the same logic leads to the units J/kg or J/m3.

Many derived units can be expressed in more than one form, but professional use usually settles on a single convention. For example, the unit of dynamic viscosity could be expressed as kg/(m•s) or N•m² or Pa•s. Only the last form is prevalent.

Holdovers From the Past

As pointed out in the letters to the editors, there are inconsistencies in the sense that non-SI units and terms remain in local (and, in some cases, general and approved) use. They mostly reflect a tradition that is slow to die.

Here are several examples of terms carried over from the past that are still in common and approved use.

The degree Celsius (symbol °C) is an alternate name for the kelvin when a temperature increment is meant. It is also a name that designates a temperature on the Celsius scale.

The degree in plane angle (symbol °) is an alternative to the SI radian.

The liter and milliliter are the everyday usage alternates for dm³ and cm³, respectively.

Prefixes (for example, kilo-, centi-, and milli-) often precede the name of a unit. Prefixes were devised to simplify, like the familiar 10ⁿ notation (for example, 50,000 expressed as 5x104). The prefixes also avoid the creation of unnecessary new units, like 5,280 feet grouped into one mile.

The 10ⁿ notation is impractical for the non-scientific person, and the creation of new units is impractical for everybody, because in the modern world it would necessitate coining thousands of names and subjecting each of us to memorizing hundreds of them.

Instead, the prefixes shorten long numbers to make them convenient for everybody to use.

The kilogram, with a cylinder of platinum-iridium alloy as its standard, is the only SI base unit still defined by an artifact.

With a rare exception, American students consider km and mm two separate units. That misconception trails through the college years among all but the most astute students and is a leading argument against metric for having "too many and long-named units." Of course, km or mm is a way of saying in shorthand "a thousand meters" or "a thousandth of a meter." They are just multiples of units.

We seem to be able to treat "kilobyte," "megabyte," and "gigabyte" quite comfortably. This author has not heard as yet anyone claiming them to be three different units.

While the SI committee has so far established 20 prefixes, far fewer—perhaps eight—are needed in daily life and in common technical work.

Ten are more than most people need.

Getting to Like SI

Most people, once they understand it, like SI for its logic, consistency, and lack of conversion factors. On the other hand, some older U.S. engineers don't like to use it. This is understandable. One dislikes anything that one does not understand and has little feel for.

For technical documentation, the preferred way of writing SI prefixes and units is by their symbols; for example, 5 kg, not 5 kilogram or five kilograms.

SI units, prefixes, and rules were established to facilitate data communication worldwide. They represent a compromise intended to suit all languages, to ease arithmetic manipulations, to prevent ambiguity, and to retain some of the tradition of the metric system.

If each symbol is written according to the SI rules— distinguishing between uppercase and lowercase letters, and between the Latin and Greek letters—it will be intelligible anywhere, regardless of the script and language a nation uses.

sidebar: seven units, defined by their keepers

Stan Jakuba is president of S I Jakub Associates in West Hartford, Conn., a firm that specializes in training and consulting in metric standards.

home | features | news update | marketplace | departments | about ME | back issues | ASME | site search

© 2001 by The American Society of Mechanical Engineers