ÐÓ°ÉÔ­´´

Historians and scientists differ profoundly in their beliefs. ÐÓ°ÉÔ­´´s
believe that patterns found in the past or uncovered by experimentation
can be predictive, that ‘laws of nature’ will follow as generalisations
from experimental results. Historians, not held to an obligation to find
laws, draw their mission from the simpler hope that a historical record
can be recreated, and that we can learn from it.

Molecular biologists are now coming to realise that human DNA is a historical
record, as well as a biological molecule subject to experimental modification.
Species have been born and died since life began. Their history is written
in their DNA. In the DNA of each of us there are passages that can be found
in every plant and animal alive today. These are ancient passages, successful
parts of the failed survival strategies of the long-gone ancestors of us
all.

Comparing the texts of DNA from different species alive today has already
begun to provide new details of the idiosyncratic history of the coming
and going of species on this planet. No previous species has had the skill
to understand its origins, and none the ability to search its origins for
clues to its future fate. As the human genome is decoded and deciphered,
we can hope to do both.

Can the unique history of life really be studied as a problem in molecular
biology? Molecular biology has been built upon the assumptions of physics,
chemistry and the other so-called ‘hard’ sciences: laws of great generality
and perfect stability do exist in nature; living things are governed by
these laws. Yet the study of natural selection stands outside this framework,
even though many students of the process begin as scientists and use the
tools of various scientific disciplines. Students of natural selection can
have no hope and no wish for eternal laws. Physicists can predict the next
solar eclipse, but no one can predict the next species. Trained as scientists
but thinking like historians, students or natural selection are pleased
to accept the contingent aspects of current and past life, the certainty
that we will not come this way again.

The study of the sequences of genomes has now drawn molecular biology
into thinking of life in this historical way. It is arguably a ‘paradigm
shift’, in the terms of the philosopher of science Thomas Kuhn. Our ability
to decipher and understand the genomes of many species, including our own,
has made molecular biologists into historians despite ourselves. As we read
out the details of our origins in the evolved differences between our DNA
and the DNAs of neighbouring species, we will begin to think more and more
like historians. We have begun to see the texts encoding individuals and
species as historical records. We should begin to approach them not as molecules
alone, but as one would a library of ancient books documenting the history
of life on this planet. The DNA texts of species alive today are full of
references to this common past, references which we first recognised by
the visible characters they encoded, and that we now can read in their original
DNA language. We have also found examples of the richness of a true historical
record: meaning in DNA is encoded in many layers, some fully predictable,
others not.

The myth of objectivity

As we shift to history, we need not surrender any of the tools of science,
nor even the methodologies of experimentation. But we will have to accept
the fact that the same species do not evolve twice. When we choose which
part of the historical record, or the human genome, to reconstruct or analyse,
we will be able to admit that our priorities are subjective, as they really
always have been. The shift will happen as we think about DNA differently,
as we come to accept the limits of history. Natural selection, the author
of our DNA texts, is a network of causal relationships so complex that we
will never break the final code linking DNA sequences to the survival of
species. We simply have to admit that, no matter how many sequences we unravel,
we will not be able to reduce natural selection to a set of predictable
laws.

This change in paradigms will also help those biologists who struggle
to understand natural selection without a full grasp of the powerful generalities
and tools molecular biology has brought to the subject. For example, historically-based
scientists assess the degree of difference between species by counting the
number of different ‘characters’ between them. Molecular biology is needed
to bring sense to the notion of characters. A ‘character’ may have any molecular
complexity. A single base-pair difference may count as a character in one
study, while another study will take as one character the difference between
chimpanzees, who lack spoken language, and humans, who possess it. Because
about four-tenths of all the messenger RNAs made from the human genome are
made specifically and only in the human brain, is is likely that such a
‘character’ as the ability to form languages has the molecular complexity
of tens of thousands of genes.

To give another example of the value of a historical perspective, a
common path to the origin of new species is the appearance of incompatibility
in mating. We would like to know which DNA sequences, if any, block the
appearance of fertile offspring in the mating of sister species such as
donkey and horse or lion and tiger. Learning this, we would then begin to
learn the extent to which the evolution of species involves chromosomal
incompatibilities, and the extent to which it involves complex genetic changes
that somehow subtly alter the behavior of animals, making them unable to
reproduce.

Perhaps the most useful place for the historical paradigm to take hold
would be within America’s federal research establishment, especially in
the multi-billion-dollar apparatus of the National Institutes of Health.
Infectious diseases are, after all, monuments to our intertwined genetic
past and present coexistence with a zoo of invisible organisms, while our
inherited diseases are entirely a matter of our own particular history as
a species.

Up till now, the clash of paradigms has not been easy. It has brought
discord and confusion to the conversations among evolutionists and molecular
biologists, and it has pushed many molecular biologists into fanciful and
unconvincing arguments for seeing the sequence of human DNA not as an historical
document, but only as a ‘database’ to guide further molecular biology. Mapping
molecular biology to natural selection does not work very well, because
it is so hard to figure out how to use the tools of the trade if there will
not be any new laws to discover. Seeing DNA as a text is the way to relieve
these tensions. No one wants to be told that a book, or a doctrine, or an
historical train of events, will predetermine their fate. Similarly, even
molecular biologists should be pleased, when they think about it, to realise
that they will not find rules and laws and codes in the functions of the
genome which might restrict the free will of their own brains.

Why is this paradigm shift important, beyond its discomforting effects
on a few of my colleagues? Because the vision of DNA as a text is closer
to reality than the vision of it as a molecule from which laws may be found.
Molecular biology, and biology generally, carry a considerable burden of
unexamined determinism. No serious historian expects the historical record
to yield rules that predict the course of history, of which order historical
or political events by their closeness to an historical ‘ideal’. Every claim
to have found such laws of history has turned out to be premature, and many
have also been excuses for terrible, murderous behaviour. Yet, many molecular
biologists continue to expect to find such laws in their work, because unlike
historians they have sometimes succeeded: the self-copying double helix
and the genetic code are examples of such powerful and fertile generalisations.

Toe use human DNA wisely, we cannot approach it looking for ways to
outwit natural selection, or for rules and codes with which to describe
an ‘ideal’ or ‘best’ human genome. The best insurance against these mistakes
is to see the genome as an historically grounded text. Then the notion of
an ideal human DNA would be seen for what historians know it to be – a mistaken
dream. As historians using the tools of molecular biology, we will be able
to appreciate the human genome, without allowing ourselves to be drawn into
empty dreams of perfection. Let us look at DNA as an interesting text, and
take both pleasures and lessons from the past by learning to read it as
clearly as we can. Each of us owns a copy that has its own particular typographical
errors and novel phrases, and in it we can find the reasons why we must
die.

Robert Pollack is professor of biological sciences at Columbia University
in New York. This article is adapted from his book Reading DNA, which will
be published by Harvard University Press next year.