Universities and research institutes shall establish procedures for dealing with allegations of scientific misconduct. They must be approved by the responsible corporate body. Taking account of relevant legal regulations including the law on disciplinary actions, they should include the following elements :

– a definition of categories of action which seriously deviate from good scientific practice and are held to be scientific misconduct, for instance the fabrication and falsification of data, plagiarism, or breach of confidence as a reviewer or superior,
– jurisdiction, rules of procedure (including rules for the burden of proof), and time limits for inquiries and investigations conducted to ascertain the facts,
– the rights of the involved parties to be heard and to discretion, and rules for the exclusion of conflicts of interest,
– sanctions depending on the seriousness of proven misconduct,
– the jurisdiction for determining sanctions.

The law on disciplinary actions legally takes precedence over these internal institutional procedures as far as sanctions touching the relationship between employer and employee are concerned. Equally, other legal regulations e.g. in labor law or in the law on academic degrees cannot be overridden by internal rules. The present recommendations are not meant to replace these existing regulations, but to call them to memory and to complement them.

Existing legal regulations do not cover all forms of possible misconduct in science, and in part they serve to protect rights other than the credibility of science and the conditions for its functioning. Owing to the different aims and contexts of these regulations, they partly postulate additional assumptions and requirements which go beyond scientific misconduct as such or address other concerns. They are not adapted to the configuration of interests typical of allegations of scientific misconduct. For instance, they do not adequately take account of the interests of the accused person, of his or her research institution, and of the 'whistle blower'. Often, legal procedures take several years.

In spite of their partly antagonistic standpoints, the person whose work has been challenged, his or her institution, and the person who has raised allegations, share an interest in a rapid clarification of the allegations and in avoiding publicity. All three wish to protect their reputation. The rules of procedure for dealing with allegations of scientific misconduct must take into account this common interest of the parties involved. They should therefore suitably provide for a procedure in several steps:

The first phase (inquiry) serves to ascertain a factual basis for judging whether or not an allegation is well founded. In this phase, the need of the respondent and the 'whistle blower' for confidentiality is balanced against the aim of reaching a clear statement of the facts within a defined short time. In this first phase, the protection of the potentially innocent respondent is particularly prominent. It ends with the decision whether the allegation has substance and therefore requires further investigations, or whether it has proved baseless.

A second phase (investigation) includes such additional inquiries as may be necessary, in particular hearings and recordings of evidence, the formal declaration that misconduct has or has not occurred, and finally the reaction to a confirmed allegation. Reactions may take the form of a settlement or arbitration, of recommendations to superiors or third parties, or of sanctions (including e.g. the obligation to retract or correct publications with proven irregularities) imposed through the authority empowered for this in the individual institution. The protection of public confidence in science requires that not only the investigation and confirmation of the facts, but also the reaction to confirmed misconduct happen within a reasonable period of time.

Such procedures, as has been noted above, reach their limits where legal regulations apply. In the first phase of inquiry, it will not always be possible to reach an exact conclusion on the precise nature of a case. The procedural character of the inquiry phase will therefore have to be measured against the requirements of related legal proceedings to ensure that findings established in this phase may, if necessary, be used in these proceedings as well.

The relationship between internal institutional procedures and legal proceedings, e.g. according to the law on disciplinary actions, is not simply a question of determining jurisdictions or competences in parallel or joint investigations. Internal regulations may, depending on the nature and the seriousness of misconduct, offer consensual solutions through conciliation or arbitration. These generally have the advantage of allowing procedures to be concluded speedily and on the basis of a settlement between the parties involved, i.e. without the judgment of a third party having to resolve the controversy. The conciliation procedure, which is obligatory according to abor law for litigation concerning employer-employee relationships, shows that consensual settlements are well adapted to the long-term character often typical for employment. To avoid an erosion of the advantages of such alternative dispute resolution through time-consuming confrontations on the person of the arbitrator and on the settlement proposed, internal regulations should prescribe time limits after which formal legal proceedings (with their specific advantages and disadvantages) shall become mandatory.

Settling a dispute on a consensual basis has a potential for peace-keeping and may in many circumstances do better justice to a case than the decision by a court of justice on the basis of abstract categorizations of the facts and their legal consequences. On the other hand, this flexibility must not lead to preferential treatment for individuals or to allegations being swept under the carpet without proper clarification.

When new procedures for conflict resolution have been instituted abroad, it has proved useful to collect data for their evaluation at a later date, e.g. in the institutions involved, from the beginning of their implementation. Such data may serve as the basis for a critical evaluation of new procedures after a pilot phase, and for their improvement.

Depending on the nature of the interventions into the rights of the parties that internal regulations allow for, their juridical character, which makes them subject to verification by the courts, has to be taken into account. Such interventions may already occur in the inquiry phase, and the imposition of concrete sanctions will certainly fall under this category.

Both phases of internal procedures, inquiry and investigation, must conform to the following principles:

a) The regulations must specify in advance who officially receives allegations of scientific misconduct, when inquiries and investigations are to be initiated, by whom, and in what form, which steps are to be taken to set up decision-making bodies, whether they be ad hoc groups or standing committees or take a mixed form, e.g. with a permanent chairperson and individually appointed members from the institution itself or from outside. Ideally the academic members of an institution should be in control of the proceedings and have the majority in the decision-making bodies. However, involving experts from outside will always serve objectivity and may be indispensable in smaller institutions.

b) Conflict of interest of a person involved in investigations must be arguable both by him- or herself and by the respondent.

c) The respondent must have a right to be heard in every phase of the proceedings.

d) Until culpable misconduct is proven, strict confidentiality must be observed concerning the parties involved as well as the findings reached.

e) The result of an investigation shall be communicated to the science organizations and journals involved at a suitable time after its conclusion.

f) The individual phases of the procedure must be concluded within appropriate time limits.

g) Proceedings and results of the individual phases must be clearly recorded in writing.The implementation of this recommendation will, as is evident from the above, require considerable juridical expertise.

–> 5. Further reading (recommended by NAS) (i)