poly: Heinleinian eugenics

From: Carl Feynman <carlf@alum.mit.edu>
Date: Thu Feb 26 1998 - 07:58:25 PST

At 01:14 AM 2/25/98 -0800, D. Brin wrote:

>In a hundred years, Robert Heinlein may be best known not for his novels
>but for the 'Heinlein Solution' to the problem of human genetic
>engineering. It can be found in his utopian novel Beyond This Horizon
>(which alas, also has a silly Code Duello with guns everywhere). In his
>future, actually CHANGING the germline is disallowed, but husband+wife are
>helped to sift ALL their sperm + ova to calculate which combination they
>want to use, thus having the best possible 'natural' child... one they
>might conceivably have had anyway. A brilliant notion... if it were
>technically possible.

Something close to it is close to possible: sperm selection. If it were
possible to sort sperm according to which genes they contained, one would
be halfway toward the Heinleinian method. It has been feasible to sort
cells according to their surface proteins for twenty-five years. It has
been feasible to sort sperm by whether they contain a Y chromosome for
about two years (sorry, no reference-- somebody at the veterinary school of
the University of Texas developed this). I would not be terribly surprised
if someone announced a method of sorting sperm according to whether they
contained a given allele, either of a particular gene of interest or of a
nearby marker gene. It's not quite as good as the Heinleinian method,
since it doesn't let you pick the egg, and it doesn't give you the ability
to select among all possible arrangements of paternal genes. In order to
have an adequate number of sperm left after the selection, you could
probably only select for a dozen or two genes at most. But compared to
methods requiiring in vitro fertilization, it is much less invasive, and
potentially far cheaper.

This development would be applauded by nearly everyone, since it could
quickly eliminate recessive genetic diseases, and do it without the
abortions inherent in the present system based on amniocentesis. After the
technology became widespread, people would start using it to select for
traits other than those usually considered disease-- things like
appearance, intelligence and longevity. By then it would be too late to
stuff the genie back in the bottle.

It is of interest to consider what traits are worth selecting for. After
some thought, here's what I came up with:

(1) The trait should have a substantial genetic component.

(2) The trait should be expressed early in life. There's no point to
selecting for a decreased chance of heart attack, since that's not a
substantial cause of mortality at ages under 50. By 50 years from now, we
will probably have developed better ways to deal with heart attacks. On
the other hand, appearance and IQ develop and childhood and are then fairly

(3) The trait should not be a positional good. (A positional good is one
that provides benefits according to where it stands in relation to the
other available goods, as opposed to any inherent characteristics.) For
example, consider basketball skill-- only the top 1000 basketball players
can have a professional career. Because there are far more good basketball
players now than there were 30 years ago, the value of any particular level
of skill has gone down. Engineering children for positional traits will
make all the unmodified kids worse off, without producing any overall
benefit. The main positional trait that is genetically controlled is
height. Taller men are far more succesful at attaining positions of
respect and authority. A rational parent might expend considerable effort
at creating a tall son. But since many other parents would make the same
decision, this might incite a race toward extreme height. Better to just
not let it get started by banning selection for height.

I wish this technology had been around before I concieved my children; then
I wouldn't have to worry about them inheriting my inflammatory bowel
disease. Right now it's not known what gene tends to cause IBD, but that
state of affairs won't last long. I recently donated a blood sample to a
project that may well have narrowed it down to a particular lymphocyte
antigen. They're going to look at my sample (and several others, of
course) and see if it confirms their theory. Progress marches on.

Received on Thu Feb 26 16:10:54 1998

This archive was generated by hypermail 2.1.8 : Tue Mar 07 2006 - 14:45:30 PST