Introduction

Genetics massively borrowed concepts from informatics. These concepts are used in genetics at two different albeit related levels.

1. At a basic level, genetics borrowed the very notion of information to explain the mechanisms of life; an example is the famous “Central Dogma of Genetics” that describes, roughly, the way in which “The transfer of information from nucleic acid to nucleic acid or from nucleic acid to proteins may be possible … but transfer from protein to protein or from protein to nucleic acid is impossible. Information means here the precise determination of sequence, either of bases in the nucleic acid and or of amino acid residues in the protein”. (Crick, 1958)

2. At a superior level, molecular biologists claim that cells are machinery similar to computers; this cell-machinery actually contains devices useful to build up unique biological beings starting from the information stored in DNA.

Different authors (i.e. Mahner and Bunge,1992; Lewontin, 1992; Marturano, 2003, and others); questioned this massive use of informational concepts; some authors claimed that such a massive and unquestioned use of concepts first lead to a coagulation of the new science of molecular biology around common basic concepts, but, later, lead to an ideological use of such concepts for business and political purposes.   

In this essay, I will first analyse some basic information-related concepts of molecular biology and then I will elucidate the ethical consequences of their misuse. It is indeed very important to understand the way the information-related concepts of molecular biology are interpreted in order to understand the reason why their incorrect application - and consequent rhetorical use by geneticists - turns into an ethical failure.

1. Genetic Information

The idea of “genetic information”; that is, genes containing an amount of information (the so-called TACG amino acids sequence) able to build a human being up, is today a seldom-challenged triviality. This idea is fundamental to the so-called “Central Dogma” of genetics. The “Central Dogma” as originally formulated by Crick is a negative hypothesis that states that information cannot flow downwards from Protein to DNA. Its complement, the “Sequence Hypothesis” is often conflated with the Central Dogma. Under it, DNA is transcribed to RNA, and RNA is translated into protein. More abstractly, information flows upward from DNA, to RNA, to proteins, and, by extension, to the cell, and, finally, to multi-cellular systems. In the ensuing years, many scientists have merged the two hypotheses and refer to them collectively as the “Central Dogma”. We will use the term in this latter collective, conjunctive sense. So this enlarged notion of “Central Dogma” or - according to Berlinski (1972) who uses the term in the sense of Kuhn - paradigm “encompasses an account of the cell’s ability to store, express,   replicate and, change information. These are the fundamental features of life, no less; and a schema that says something interesting about them all has at least a scope to commend it” (Berlinski, cit.)  

Readers that are more expert will understand the characterization of the “Central Dogma” as based on the so-called broadcasting theory of communication in which we have just only one information sender and multiple information recipients, and information flows one way from a receiver to recipients.   

However, the idea of genetic information, or better, of a code script into the cell is ascribed to Erwin Schroedinger; the code script would be “a sort of cellular amanuensis, set to record the gross and microscopic features of the parental cell and pass the information thus obtained to the cell’s descendant (Schroedinger, 1944).

Other authors reject the idea that the concept of information does apply to DNA because it presupposes a genuine information system, which is composed of a coder, a transmitter, a receiver, a decoder, and an information channel in between. No such components are apparent in a chemical system (Apter and Wolpert, 1965). Even if there were such a thing as information transmission between molecules, this transmission would be nearly noiseless (i.e. substantially non-random), so that the concept of probability, central, to the theory of information, does not apply to this kind of alleged information transfer (Mahner and Bunge, 1997).           

2. A semantic or a syntactic theory of genetic information?

Several authors argued that molecular biology developed at the same time as computer technology and information theory; these two parallel processes have remained parallel. The biological notion of “information” has developed independently from the one advance by Shannon (in computer science). The expression “genetic information”, used for the first time in Watson and Crick (1953), has a metaphorical connotation – as we have seen before - without any particular reference to the nature of “code”. As Crick explained later, for information they intended the specification of the amino acids sequence in the proteins; in Crick’s mind such a notion, so to speak, was an “instructive” one (see Fox Keller 1995) rather than “selective” as it is in Shannon’s theory.

In the 1950s the biological notion of information was identified with the notion of “specificity”. In that context, specificity and information had become synonymous terms: they were based on the concept of uniqueness of the sequence as a condition for auto replication. According to Corbellini (1998), this idea illustrates a surprising analogy with a particular approach to the notion of computational information understood at the end of the 1950s. Corbellini also believes that this model overcomes some applicative limits of Shannon’s original model, with respect to the importance of the meaning attached to transmitted information. Bosnack (1961) proposed a model in which information was interpreted as “specificity” that is the difference between two entropies (the one before choosing a particular message and the one after such a choice) within a repertoire. Applying such a mathematical definition of specificity, Bosnack, was able to overcome the subjective connotation implied at the semantical level of Shannon’s original notion of information. The importance of the semantical level in the biological notion of information, according to Corbellini (1998), was underlined by E.H. Hutten who saws in such notion of specificity an adequate formal definition to explain the transmission of genetic information from nucleic acid to proteins, which has encapsulated the meaning of an order. In other words, while Shannon’s idea of information has a statistical value, in biology the semantical aspect of communicative interactions plays a more fundamental role. During the 1960s – despite these characterisations of the nature of information in terms of specificity – a metaphorical notion of information become fully absorbed into the vocabulary of molecular biology, whether in the context of the nucleic acids or in that of proteins. But it was not until 1977 that robust and generally applicable sequencing methods were developed, and even then the modern bioinformatics techniques of gene discovery was still years away. Although the development of information/processing by computers proceeded contemporaneously with progress in research into biological and biochemical information processing, the trajectories of these two initiatives were never unified even if they sometimes overlapped at various points.

According to Castells, such a theoretical convergence between information and genetics technological fields has now been realised (2001: 164). This is also because modern science relies largely on computer simulations, computational models and computational analyses of large data sets. Although genetics is considered to be a process that is entirely independent from microelectronics, is not really so independent. First, Castells argues that genetics technologies are obviously information technologies, since they are focused on the decoding and eventually reprogramming of DNA (the information code of living matter). And, more importantly, “Without massive computing power and the simulation capacity provided by advanced software, the HGP, would not have been completed – nor would scientists be able to identify specific functions and the locations of specific genes” (ibid.). Sir John Sulston seems to agree with Castells: “The future of biology is strongly tied to that of bioinformatics, a field of research that collects all sort of biological data, tried to make sense of living organisms in their entirety and then make predictions” (Sulston, 2002).

In addition to this, some philosophical problems arise from the view that DNA and the Human Genome are pure informational concepts. In one sense, the convergence between the biological and computing might be thought to be associated with the massive use of computer technologies in biology. But, as Holdsworth (1999) suggests, “It is not just that computer tools are rather convenient for doing genomics and protein sequencing. Rather these… disciplines have re-organised themselves around the bioinformatics paradigm”, and doing that they have absorbed Shannon’s notion of information.

Others have argued that there is an intrinsically special value in genetic information, which is said to differ either from (a) other kinds of health information, or, (b) other kinds of information in general. According to Holm (1999), there is little support for the claim that genetic information has one or more special features that distinguish it from other health related information in any morally relevant way. The idea that genetic information has a special relevance is linked to the claims of “genetic essentialism”. Genetic essentialism can be expressed in the statement that “we are our genes”, or as Walter Gilbert metaphorically putted it: “once we will be able to pull a CD out of one’s pocket and say, ‘Here’s a human being: it’s me!”’. Genetic information would be, therefore, information about the very essence of a person, whereas other non-genetic information would be only about accidental attributes. We would still need to be able to distinguish between genetic and non-genetic information, but if we could do that, genetic information would surely be special. Genetic essentialism is widely attacked (i.e. Lewontin, 1991) even this view has influenced the public perception of genetics (Nelkin and Lindee 1995).According to Lewontin (1991: 63), however, it takes more than DNA to make a living organism and its history. A living organism at the very moment of its life is the unique consequence of a developmental history that results from the interaction of and determination by internal (genetics) and external (environmental) forces. Such external forces are themselves partly a consequence of the activities of the organism itself, produced by the conditions of its own existence. Reciprocally, the internal forces are not autonomous, but act in response to the external. Part of the internal chemical machinery of a cell is manufactured only when external conditions demand it. Therefore, genetic essentialism, which assumes the uniqueness and independence of genetic information, does not give us a plausible argument for treating genetic information in a special category. Another claim that genetic information is special compared with other kinds of health-related information is sometimes based on a further claim there is some other of genetic information that makes it different. Some have pointed out that genetic information is predictive; but it is also worth to pointing out that, on the contrary, a lot of genetic information is non-predictive and much of non-genetic health related information is predictive. Knowing, for example, the LDL-cholesterol level in the blood of an individual can also predict that person’s risk of coronary heart disease. According to Holdsworth (1999), the convergence of the computational notions of information with the biological notion enables us to see that there really are not two different kinds of information. In each of these two contexts, we find signals that are expressed by orderings of the states of physical substrates. Viewed in this light, the nature of substrates, whether silicon or carbon is irrelevant; and it is irrelevant even if, in real-world situations, there seem to be contingent reasons for drawing differences.

According to Maynard-Smith the word “information” is used in two different contexts, “It may be used without semantic implication; for example, we may say that the form of a cloud provides information about whether it will rain. In such cases, no one would think that the cloud had the shape it did it provided information. In contrast, a weather forecast contains information about weather it will rain, and it has the form it does because it conveys that information. The difference can be expressed by saying that the forecast has intentionality whereas the cloud does not” (Maynard Smith, cit.). The notion of information as it is used in biology, he argues, “is of the former kind: it implies intentionality. It is for this reason, that we speak of genes carrying information during development and of environmental fluctuation not doing so” (id.). How, then, can a genome be said to have intentionality? Maynard Smith claims that “the genome is as it is because of million of years of selection, favouring those genomes that cause the development of organisms able to survive in a given environment. As a result, the genome has the base sequence it does because it generates an adapted organism. It is in this sense that genomes have intentionality” (id.).    

3. The Cell as computer machinery

2.1. Berlinski: Bacterial cell as automata

Berlinski (cit.), explores another interesting analogy between languages and the genetic code (Wendell-Waechtler and Levy, 1973). More interestingly, he wishes to identify a mechanical instrument, which would “pass” for coding into DNA all and only those well formed formulae and rules out those which are not. The literature of mathematical linguistics, according to Wendell-Waechtler and Levy (id.) would be a species of potentially infinite automaton or Pushing-Down Storage Machines.

Berlinski, indeed, suggests that concepts such as code, information, language, control and regulation and translation come easy to the biologist as he describes the cell according to lights provided by the central dogma. These, according to Berlinski (cit.), are distinctly theoretical notions, and researchers have the laudable ambition of setting biological entities in a context amenable to abstract description and analysis. Such settings would provide a formal model for the bacterial cell. But, code theory says nothing about the organisation that may be in evidence in the nucleic acid; this deals with problems that arise when information is passed along a channel of communication. Here stops the usefulness of information theory. It draws a difference between the informational macro-molecules (the nucleotides and the proteins) and the other cellular constituents; according to Berlinski, automata theory would be “the source for the most natural models of the bacterial cell” (id.).     

In its more general form automata, theory treats of the conversion of inputs to outputs via a device that admits of states. The conversion is effected between items that are discrete and combinatorial: numerals, words, sentences, and letters. The bacterial cell seems similar to this sort of treatment: sequences of nucleotides resemble sentences: codons, and words. These very intricate processes of control and organisation, Berlinski (id.) claims, evoke computing machinery of various kinds.

This idealisation of a cell thus appears as automation: the associated programs are designed to push the machine through movements that look vaguely biological. Often the machines turn out to have very strong computational capabilities. The idea that a cell is a machine is attractive; J. Monod, for example, has confessed that it is just in the discovery of the machinelike nature of cell that modern biology has had its most impressive triumphs (Monod, 1971, ch. 4).

In particular, Berlinski (id.) claims, cells are somewhat potentially infinite machines; that is pushing down storage automata (PDSA). A PDSA is a machine, which accept or reject input strings: a given string is accepted if the machine reads it when the stack is empty.   This treatment, as readers can see, is very close to Turing machines. The bacterial cell, Berlinski continues, suffers some idealization when set as a PDSA: bacterial systems quite obviously must complete their computations within close limits of time and space, and in any case specimens of, for example, Escherichia Coli have boundlessly large memories at their disposal. Even though constructing a biological automaton would be quite difficult, biological PDSA embody algorithms for the conversion of codons into string of nucleotides. They thus control the nature of proteins that are sequenced and the order in which they are synthesized. Therefore, states of the bacterial cell can be identified with the states of a PDSA; the full set of codons formed from a nucleotide alphabet, with its set of input symbols.    

An automata-theoretic approach satisfied many algorithmic intuitions about the cell: the feeling that the growth and regulation of the cellular machinery is basically a recursive process in which a finite set of elements are teased into a complex construction: a class of machines with fixed properties and limitations but no specific computational powers. But, the central dogma requires something quite different: a system in which elements of the base vocabulary change at random and thus form output sequences that wholly unlike any that the machine is prepared to manage. Computability, indeed, calls to mind procedures that are basically mechanical; but it also evokes the existence of constraints on the possibilities involved in converting inputs into outputs. The bacterial cell, with its exquisitely refined biochemical mechanism, could hardly be under the control of an object inherently liable to a form of babbling.

Finally, Berlinski (cit.), suggests to replace the central dogma with a doctrine that brings changes of biological information within the compass of automata-theoretic methods. Biological PDSA modelling, he claims, “the bacterial cell could emerge as machines that manoeuvre through only a limited repertoire of genetic re-assemblies. Such devices would instantiate algorithms that fixed in advance those strings of nucleotides which are accessible to a given organism” (Berliniski, id.). He rightly suggests that there would be few trouble in preserving the backbones of the central dogma in such machines: mutations could continue to comprise a set of equally probable events just so long as they were confirmed by constraints fixed on admissible strings of codons. Changes that resulted in strings that could not be generated would simply not arise; or, failing that, would arise without effective genetic expression.

That kind of biological automata would be marvellous machines sorting out among the proteins and instantiating a definition of life itself, but have no explanation within the neo-darwinist framework: such automata do not suddenly acquire great computational powers; the contrary hypothesis that life began with a set of cells fully committed to just the viable protein has little save elegance to commend it.

Berlinski concludes that general reflections on the introduction of automata do little to impede the unravelling of the central dogma. The original invocation of automata as models for the bacterial cell carried with it is a conceptual baggage of thought that biological automata not only arranged the affairs of the cell, but also, fixed in recursive form computational powers that were sufficient to segregate the viable from the unviable proteins.             

2.2. Maynard-Smith: Eggs as computer machineries

In developmental genetics, the model of information has moved even more far. John Maynard-Smith, one of the champions of this branch of molecular biology, starting from Mahner’sand Bunge criticism, expressed the idea that cells are actually a kind of information systems. Maynard-Smith (2000) claims that indeed there are such things like coder, transmitter, receiver, coder, or information channel; “if there is ‘information’ in DNA, copied to RNA, how did it get there?” (idem). Maynard-Smith concludes that natural selection plays the role of the coder in biological systems: “In human speech, the first ‘coder’ is the person who converts a meaning into a string of phonemes… In biology, the coder is natural selection” (idem). What of the claim that a chemical process is not a signal that carries a message? Maynard-Smith argues that this hypothesis is false because of the idea that the same information can be transmitted by different carriers, or better, that information is carrier-free, so information can be carried by chemical systems. Finally, Maynard-Smith, rejects the idea that probability does not play a fundamental role in biology just on the basis that genetics transmission of information is virtually noiseless. Rather, he argues, difficulties in applying information theory to genetics arise principally not in the transmission of information, but in its meaning.

Maynard-Smith (idem), claims that the analogy between the genetic code and human-designed codes is too close to require justification, and he points our attention out to the fact that the genetic code is symbolic but we have machinery in the cell to process information; there are a decoding machinery (i.e. tRNA), a translating machinery (ribosomes, tRNAs etc) and finally genes that contains coded information.

Finally, he claims “Yolk is just a store of nutrients: it no more carries information than the petrol in your petrol tank… An egg must also contain the machinery – ribosomes etc.- needed to translate the genetic message. The machinery is provided by the mother, and coded for by her genes. It is perhaps the classic example of the chicken and the egg paradox: no coding machinery without genes, and no genes without coding machinery” (Maynard Smith, 2000b). “What is inherited is not the dark pigment itself, but the genetic machinery causing it to appear in response to sunlight” (Maynard Smith, 2000). That arise a new analogy: an egg is similar to a computer on the respect that it contains all the machinery able to process that information useful to build up a new individuals with his/her characteristics. The informational metaphor is thus expanded: not only we have “genetic information”, but we also can talk of cells as “computational machines” in which the role of each computational element is defined.       

4. Use and Misuse of models

We have just seen two ways in which the informational and computer models were imported by philosophers of biology into the biological and genetic realm. However, these models were widely used and even stressed to other biologists such as W. Gilbert making that a common sense. Ordinary use of the model, today, has unfortunately collapsed the distinction between a model and the theory for which it is a model. There is, indeed, a one-one correlation between the propositions of the theory and those of the model; propositions, which are logical consequences of propositions of the theory, have correlates in the model, which are logical consequences of the correlates in the model of these latter propositions in the theory and vice versa. But the theory and the model have different epistemological structures: in the model, the logically prior premises determine the meaning of the terms occurring in the representation of the calculus of the conclusions; in the theory the logically posterior consequences determine the meaning of the theoretical occurring in the representation in the calculus of the premises. The widely used realist vocabulary has collapsed this fundamental distinction turning the informational model into the theory, or, even worse – as in the Gilbert example - into the actual ontological base of genetics.    

The idea that models were a dangerous tool is a well-known topic in philosophy of science. According to R.B. Braithwaite (1953, p. 93 and foll.) there are two dangers in the use of models:

“The first danger is that the theory will be identified with a model for it, so that the objects with which the model is concerned – the model-interpretation of the theoretical terms … of the theory’s calculus – will be supposed actually to be the same as the theoretical concepts of the theory. To these theoretical concepts will then be attributed properties which belong to the objects of a model but which are irrelevant to the similarity in the formal structure, which is all that is required of the relationship of model to the theory… Thinking of scientific theories by means of models is always as-if thinking; hydrogen atoms behave (in certain respects) as if they were solar systems each with an electronic planet revolving round a protonic sun. But hydrogen atoms are not solar systems; it is only useful to think of them as if they were such systems if one remembers all the time that they are not. The price of employment of models is eternal vigilance” (Braithwaite, cit., p. 93).

In much as such sense, the notion of information in biology was a fundamental operational instrument (that is an as-if thinking, according to Braithwaite) that helped and even boosted genetic research. Unfortunately, at a point, it “collapsed” or was “naturalised” (or - in Braithwaite’s own terminology - it led to an identification between the model-interpretation of theoretical terms and the theoretical concepts of the theory) reducing a quite powerful heuristic model into the very research object.

Again, for Braithwaite, there is “a second danger inherent to the use of models, a danger which is more subtle that that of projecting onto the concepts of the theory some of the empirical features of the objects of the model. This danger is that of transferring the logical necessity of some of the features of the chosen model onto the theory, and thus of supposing, wrongly, that the theory or parts of the theory, have a logical necessity which is in fact fictitious” (Braithwaite, cit., p. 94).   

According to Buiatti (1998), those who defended that model have even turned it into an untouchable dogma (Lewontin, 1992, for example, says that “Molecular Biology is now a religion, and molecular biologists are its prophets”); so that, biological organisms were reduced to living computers, in particular to the way computers were understood in the early ‘60s; and, therefore, transferring the logical necessity of some of the features of the chosen model into the theory, as Braithwaite warned.

There are many reasons why the information model in genetics was “ontologized”; from one hand, it has provided a powerful research strategy – a kind of reference guide - that has grounded and organised a new discipline: molecular biology. From the other hand, it provided a useful ideological tool for scientists to fund major research programs up such as the Human Genome Project (HGP).

This raises a moral question about whether molecular biologists used the “collapsed model” in a correct (or honest) way for fund raising or, as Lewontin seems to suggest, it was rather an ideological weapon to monopolising media interests, and capitalise on future patent rights. More importantly, media drum banging around the HGP might steal room and funds to research with a narrow focus and thus interesting just for a smaller audience. According to Vicedo, “the first task for a moral point of view is for the scientists to inform society about the development of the initiative and its implications. Lack of (or – I would add – biases in) information always raises suspicion, and leads to misunderstanding… (Geneticists should) to assess realistically the value of the project, and to avoid making empty promises” (Vicedo, 1992).

According to Vicedo (1992), one of the main problems, arising at the beginning of the HGP was ensuring the co-ordination of the different tasks and the co-operation among all the research groups. She points out: “Some regulatory guidelines could be established to secure the smooth functioning of the project, but the scientists concerned hold different view on this issue. J. Watson, for example, thinks that the groups will develop rules to co-ordinate their efforts as the investigations proceed. Other researchers, like Walter Gilbert (Harvard), think that clear rules should provide all participating members access to the results. Others suggest that the need for groups to communicate to obtain mutual benefits will force them to co-operate”. Elke Jordan believes that the HGP’s goals will be unattainable unless it is “built on teamwork, networking and collaboration”. In his opinion, “This makes sharing and co-operation an ethical imperative”. As Vicedo’s remarks suggest, co-operation was a fundamental concern since the beginning of the HGP. One cause for concern arose because of the so-called emerging patenting-and-publish system between researchers and backed by the pharmaceutical and biotechnologies industries. This factor influenced the merging of scientific research with business interests. Although debate and discussions continue, a large biotechnology industry funded by a massive infusion of venture capital and an equally significant amount of capital from large, often multinational pharmaceutical companies has become an established force (Rabinow 2000: 3).

References:

Apter, M.J. and Wolpert, L. (1965): “Cybernetics and Development I: Information Theory”, Journal of Theoretical Biology, 8: 244-257.

Berlinski, D. (1972): “Philosophical Aspects of Molecular Biology”, Journal of Philosophy, 12: 319-335.

Bosnack, F. (1961): Information, thermodinamique, vie et pensée. Gauthiers-Villars, Paris.

Braithwaite, R.B. (1953): Scientific Explanation. The Tanner Lectures. Cambridge: Cambridge University Press.

Buiatti, M. (1998): “L’analogia informatica del ‘dogma centrale’ e le conoscenze attuali in biologia”, in B. Continenza, and E. Gagliasso, editors, L’Informazione nelle Scienze dalla Vita, Milano, FrancoAngeli: 100-117.

Castells, M. (2001): Informationalism and the Network Society, In P. Himanen, editor, The Hacker Ethic and the Spirit of the information Age. London, Vintage: 155–178

Corbellini, G. (1998): La definizione informazionale della specificita` biologica. In B. Continenza, and E. Gagliasso, editors, cit.: 66–99.

Crick, F. (1958): “Central Dogma of molecular biology”, Nature, 227: 561-563

Cukier, K. (2003): Open Source Biotech. Can a Non-Propietary Approach to Intellectual Property Work in the Life Sciences, The Acumen Journal of Life Sciences, 1(3).

Fox Keller,E. (1995): Refiguring Life: Metaphors of Twentieth Century Biology, The Welleck Lectures. New York, Columbia University Press,.

Holdsworth, D. (1999): “The Ethics of the 21st Century Bioinformatics: Ethical Implications of the Vanishing Distinction between Biological Information and Other Information”. In A.K. Thompson and R. Chadwick, editors, Genetic Information. Acquisition, Access and Control, New York, Kluwer/Plenum: 85–98.

Holm, S. (1999): “There is Nothing Special about Genetic Information”, In A.K. Thompson, and R. Chadwick, editors, cit.: 97-104.

Koerner, B. (2003): “Attacking Venter Capitalism”. Wired Magazine, 6 (6),

Lewontin, R. (1992): “The Dream of the Human Genome”, The New York Review of Books, May 28.  

Lewontin, R. (1999): The Doctrine of DNA, Biology as Ideology. London, Penguin.

Lewontin, R. (2000): It Ain’t Necessarily So. London, Granta.

Marturano, A. (2003): “Molecular Biologists as Hackers of Human Data: Rethinking IPR for Bioinformatics Research”.   Journal of Information, Communication & Ethics in Society, 1 (4), 207–216,

Mahner, M. and Bunge, M. (1997): Foundations of Biophilosophy, Berlin, Heidelberg.

Maynard-Smith, John (2000): “The Concept of Information in Biology”, Philosophy of Science, 67: 177-194.

Maynard-Smith, John (2000b): “Reply to Commentaries”, Philosophy of Science, 67: 214-218.

Monod, J.(1972): Chance and Necessity, London, Vintage

Nelkin and Lindee 1995: The DNA Mystique: The Gene as a Cultural Icon, New York, W.H. Freeman and co.

Rabinow, P. (2000): French DNA. Trouble in Purgatory. Chicago, Chicago University Press.

Stallman, R. (1994): “Why Software Should Not Have Owners”, 1994. Accessed from http://www.gnu.org./philosophy/whyfree.html 03 April 2003.

Schroedinger, E. (1944): What is Life. The Physical Aspect of the Living Cell, Cambridge, Cambridge University press.

Sulston, J. (2002): “Heritage of Humanity”. Le Monde Diplomatique. December, 2002. Accessed form http://mondediplo.com/2002/12/15genome 15 February 2004.

Vicedo, M. (1992): “The Human Genome Project: Towards an Analysis of the Empirical, Ethical and Conceptual Issues Involved”. Biology and Philosophy, 7: 255–277,

Watson J.D. and Crick, F. (1953): “Molecular Structure of Nucleic Acids”. Nature 171: 737–738

Wendell-Waechtler and Levy (1973): “More Philosophical Aspects of Molecular Biology”, Philosophy of Science , 2: 180-186.

The present essay is a development of my research presented to the XV Internordic Philosophical Symposium, Helsinki, May 13-15 2004 and the 8 th Annual Ethics & Technology Conference, Saint Louis University, 24-25 June 2005. I am indebted to the remarks my colleagues made during these two meetings.   An expanded version of this paper was submitted to Tavani and Himma (eds.), Handbook of Information and Computer Ethics, Wiley, Boston, 2007