Nature © Macmillan Publishers Ltd 1997

had been conducted, then what objectionsmight be made to the limited application ofcloning to overcome extreme cases ofinfertility? Kahn asserts, first, that theparental relationship would be “entirelynew”. Not so. We have many models ofparenting: adoption, fostering, steprelationships.... All may draw on the modelof a two-parent nuclear family — in itself arelatively modern phenomenon. So, withcloning, there might be new aspects to theparental relationship. To argue that it wouldbe “entirely new” is a piece of unjustifiedrhetoric: studies of new reproductivetechniques suggest that families have littledifficulty in assimilating them intotraditional patterns (see, for example, J.Edwards et al. Technologies of Procreation:Kinship in The Age of Assisted Procreation,Manchester Univ. Press, 1993).

The second argument is that “anindividual must never be used exclusively asa means”. Kahn makes great play of the forceof “exclusively”. Perhaps he could explainwhy the cloning of a child for a man withtotal germ cell failure results in that childbeing used “exclusively as a means”. Howdoes that case differ from that of childrenresulting from medically assistedreproduction? The latter accounted in theearly 1990s for almost 12 per cent of allbirths in France.David ShapiroNuffield Council on Bioethics,28 Bedford Square,London WC1B 3EG, UK

correspondence

Sir — Axel Kahn complains (Nature 388,320; 1997) that his argument that“application to man of asexualreproduction and cloning represents anaffront to human dignity” has beenmisreported by Nature and by John Harris’sletter. He usefully restates his argument.This remains confused, like the all-too-influential French national committeereport, which he helped to draft. It isimportant to try to cast light on theconfusion.

Kahn ascribes autonomy to “theindeterminability of the individual withrespect to external human will”. He thenasserts that “asexual reproduction wouldlead to... people whose bodily form andgenetic make-up would be exactly asdecided by other humans”. This is aninteresting example of the geneticdeterminism so rightly condemned by theFrench national committee report. Couldeven so distinguished a molecular biologistas Kahn actually predict in advance theadult height and body weight of any givenclone? And, even if he could, why shouldthat affect the autonomy of the clone’sbehaviour? After all, he rightly argues thatthe genetic make-up of a sexuallyreproduced person does not affect theautonomy of that person’s behaviour.

Kahn then asserts that the relationshipbetween ‘creator’ and ‘created’ would be“entirely new”. But why, and in whatrespects? He further asserts that this “hasobvious implications for human dignity”.

Again, why? At a Unesco meeting in May,another contributor to the French nationalcommittee report argued that cloningmight lead to the creation of second-classcitizens and even to a revival of slavery. Theanswer to that rhetoric is that humans, alas,have been well capable of creating second-class citizens and slaves without any resortto the technology of cloning.

Could we now have a philosophicallyserious debate about the only morallyappropriate proposed use of humancloning, namely Robert Winston’ssuggestion (Br. Med. J. 314, 913–914; 1997)that it might provide reproductivepossibilities “for those men who exhibittotal germ cell failure”? Winston points outthat “even if straight cloning techniqueswere used, the mother would contributeimportant constituents — hermitochondrial genes, intrauterineinfluences, and subsequent nurture”.

Let us make two assumptions: first, thatin any society where this was thoughtappropriate, such cloning would be tightlyregulated. Second, that many years ofanimal trials would have established thathuman clones were not likely to be at unduephysiological risks compared with sexuallyreproduced humans. At this point, manywould argue that the animal experimentswould not be justified. If that argumentwere accepted, then, on those groundsalone, it would be right to rule out humancloning.

If, however, the necessary animal trials

NATURE | VOL 388 | 7 AUGUST 1997 511

Cloning, dignity and ethical reasoning

Alphabetical ordersSir — Recent correspondents1,2 havesuggested that authors with initials earlierin the alphabet are more likely to be citedsimply because they are more numerous, afact borne out by examination of thedistribution of surnames in the Londontelephone book.

However, comparing the citation rate of each letter with the number oftelephone users of each letter is an inferior attempt to control for differencesin the representation of initials to the

method employed in the original study. The correlation originally identified3

was between the number of citations foreach paper published by authors of aparticular initial and their surname’salphabetic position. This controls for theuneven distribution of initials amongauthors — if Zs are rare, they will publishfewer papers.

Perhaps the crude technique of usingLondoners as a control removes thecorrelation because only a subset of papersare published by people with Englishnames. Clearly there is still a case toanswer, and given the growing importanceof citation indices, one deserving ofserious research.Tom Tregenza Department of Biology, University of Leeds,Leeds LS2 9JT, UK e-mail: gentbt@leeds.ac.uk

1. Shevlin, M. & Davies, N. O. Nature 388, 14 (1997)

2. Jaekel, K. M. Nature 386, 643 (1997)

3. Tregenza, T. Nature 385, 480 (1997)

Back to basicsSir — Misleading references in scientificpapers are common where a research grouphas studied the same topic for many years.

I recently gave a course on the energymetabolism of Escherichia coli. In a paperpublished in 1996, the authors say: “For ß-galactosidase assay, cells were grown in aglucose-containing minimal medium (pH 7.0) unless otherwise indicated4”.

Reference 4 is a paper from the sameresearch group published in 1990. In thispaper the sentence reads: “For the ß-galactosidase assay, cells were grown inglucose (40 mM) minimal medium (pH 7.0) (ref. 13) unless otherwiseindicated”.

Reference 13, published in 1985, reads:“For galactokinase assays and mRNAisolation, cells were grown in 50-mlvolumes of glucose minimal medium32

supplemented with the indicated electronacceptor”.

Finally, the composition of the

medium is given in reference 32.This is a random example of a very bad

habit. But why don’t authors give thecorrect reference in every paper?Jukka HeinonenDepartment of Biochemistry, University of Turku,FIN-20014 Turku, Finlande-mail: jukahe@utu.fi