An engineering definition of design

Bill Hamilton (
Thu, 23 Apr 1998 21:38:23 -0400

Howard has been -- rightly -- pressing the ID advocates for a definition of
design. I hope they will respond to his entreaties.

In the meantime, I'd like to offer a few thoughts on design from an
engineer's point of view.

When an engineer designs an object, he begins with definitions of what the
object is supposed to do and/or what it is supposed to look like, and a set
of constraints such as cost, strength of materials considerations,
dynamical considerations, etc. I will call the set of statements of
desired form and function and constraints the specifications. So, from an
engineering standpoint, a designed object is one whose form and/or function
conform to a set of specifications. Incidentally, considerable thought may
go into producing the specifications themselves. The specifications
themselves could probably be considered a designed object. (I will use
"specifications" and "design rules" interchangeably in what follows)

This definition can be applied to objects other than engineering designs.
Anything whose form and/or function conforms to a sufficiently complex set
of rules (or specifications) could be considered designed by this
criterion. Examples might include

A painting.

The artist wants to depict a scene or a design that conveys a message or a
feeling, and conforms to the artist's views about proportion, balance,
color combinations, etc.

A musical composition

The composer wants to communicate a feeling while observing cultural rules
about rhythm, harmony, tempo, etc.

A poem

The poet wants to convey a message, frequently with strong emotional
content, while observing rules about rhyme, meter, etc.

An automobile

The designers want to create a form that says, "take me home with you"
while satisfying constraints that enable the car to be built for acceptable
cost, to be safe, to meet fuel economy and emission requirements.

A sentence

The speaker wants to communicate information while observing rules about
grammar, syntax, usage, etc.

One of the questions we've been wrestling with -- off and on over the past
few years -- is the question of how you identify a designed object. _If_
the definition of design I have given above is accepted, then the question
of identifying design can be fairly straightforward. If I have a set of
design rules and an object that meets them, and the number of choices
required to ensure conformance is nontrivial, then I can reasonably infer
that the object is designed.

But I'm not convinced that this is of much help, because you must know the
design rules. In the examples Bill Dembski gives, the observer must know
the design rules to identify design. Chinese ceases to sound like
gibberish _if_ you know Chinese -- the design rules for a language called
Chinese. The psychologist can decide that the rat has learned the maze
_if_ he has a map of the maze -- the design rules for traversing the maze.
In the "Methinks it is like a weasel" example, the underlying sentence must
be in a language you recognize. Had the underlying sentence been in
Sanskrit (transliterated into the English alphabet of course) , the fact
that it was written in a Caesar cipher wouldn't have helped much.

Of course the ID advocates may have a definition of design in mind which
points to a different means of identifying design that doesn't require
design rules. But they will have to tell us what that definition is if
they want to communicate their insights.

Bill Hamilton
William E. Hamilton, Jr., Ph.D.
1346 W. Fairview Lane
Rochester, MI 48306
(810) 652 4148