I am very pleased to have the opportunity to respond to the letters of Drs Blackwell and Aitchison, and Professor Sims. They are all relevant to the important general issue of whether authors of papers on topics known to be controversial (such as abortion and ethnicity) should always be obliged to state their own background position in full. I suggest that the answer to this must always be ‘Yes, definitely.’

In scientific research, all possible attempts should be made to keep biases to a minimum, but unavoidable human influences can still be there and need to be known by readers if they are to understand both the data and the conclusions. These include the reasons for the research or review, the conclusions of any previous related studies by the same authors, possible biases in the methods of collection and analysis of the data, and possible biases in the conclusions of the authors. Different readers may then interpret the findings in different ways, depending upon their own viewpoint. If authors of papers on controversial topics follow these guidelines, and always state whether their conclusions are based solely upon the data of the study or also upon other background personal reasons, then the question of ‘outing’ will never arise. Similarly, on this line of reasoning, the simple statement of undisputed facts should not be regarded as ‘an attack’. There is wide agreement that financial rewards in the background must always be declared, so surely the same should apply to other potentially biasing influences.

Professor Sims's reference to the ‘psychiatry of the 1960s’ puzzles me, and without specific examples I cannot comment on this.

The overall point at issue is that readers should be able to make up their own minds, and not be limited only to what the authors believe to be the best interpretation of the study. This may be rather perfectionistadvice, butatleast it givesamodelasatarget.

A more specific issue relates directly to the paper by Fergusson et al ^{Reference Fergusson, Horwood and Boden1} and to the comments by Professor Casey suggesting that this study constitutes evidence that special emphasis on the potential psychiatric hazards of abortion should be an obligatory part of psychiatric educational programmes.

Drs Rowlands & Guthrie ^{Reference Rowlands and Guthrie2} seem to me to give a good summary of this whole problem: ‘Whether abortion causes harm to women's mental health is a question that is not scientifically testable, as women with unwanted pregnancies cannot be randomly assigned to abortion v. abortion denied groups. It seems inappropriate therefore for Casey to talk of potential litigation against abortion providers for failing to provide information on a possible causal link between abortion and subsequent mental health problems.’

Debates on this topic and others such as racism tend to be endless, so I suggest that if anyone wishes to continue further, they should do so by direct personal emails.