I Sveriges kriminalvårdsanstalter, häkten och polisarrester sitter inte bara personer som är misstänkta eller dömda för brott, utan även ett ganska stort antal utlänningar som är tagna i förvar. Dessa utlänningar är inte frihetsberövade på grund av att de har begått något brott, men de är ändå intagna i institutioner som annars enbart är avsedda för personer som är misstänkta eller dömda för brott. Anledningen till att utlänningarna är placerade här är att Migrationsverket ansett att de inte är lämpliga att vistas i Migrationsverkets förvar, där förvarstagna normalt placeras. Internationella organ som har att övervaka skyddet för mänskliga rättigheter har riktat kritik mot Sverige för att förvarstagna placeras i kriminalvård. I följande tvådelade artikelserie analyseras lagregleringen för placeringen av förvarstagna i kriminalvård. Här framförs också ett antal kritiska synpunkter på reglernas utformning och tillämpning. I denna första artikel kommer jag att behandla grunderna för placering i kriminalvård. I en andra artikel kommer jag att behandla frågor om hur de som placeras i kriminalvård ska behandlas samt vissa processuella frågor

This doctoral dissertation studies the legal standard of proof called reasonable suspicion (skälig misstanke). This standard is central for criminal investigations as the majority of state criminal procedural actions (tvångsmedel) are tied to it, as well as the majority of a suspect’s rights. Despite the central role played by this standard, it has not to any significant extent been researched previously. What the standard entails cannot be seen from either the law or legislative history, and there are no published cases discussing it. As the meaning of the standard is not clear, there is much room for discretion in assessments as to whether the standard has been met. There consequently is a need to clarify what reasonable suspicion entails. What is required to fulfill the standard? How is reasonable suspicion to be assessed? These questions are answered in this dissertation.

My investigation demonstrates that reasonable suspicion is a three-pronged standard:

A requirement of preponderance of the evidence

Reasonable suspicion entails a requirement of a preponderance of the evidence (sannolikhetsövervikt). This means, somewhat simplified, that it should be more probable that the suspect is guilty than innocent. The hypothesis underlying the reasonable suspicion of alleged guilt (gärningshypotesen) should be more probable than all other plausible hypotheses combined.

A requirement of individualization

The evidence must be sufficiently individualized in order to reach the level of reasonable suspicion. The requirement of individualization is in place to prevent that persons are suspected on the basis of group membership. The same is true with respect to persons belonging to groups where the probability of crime is generally high. In such situations, there must be a concrete circumstance pointing to just that person committing a certain crime. It can be, however, circumstances having very low evidentiary value. Group membership can still be of significance with respect to the suspect, but reasonable suspicion cannot be solely based on group membership. A generally applicable concept as to that distinguishing the general from the individual cannot however be articulated. All evidentiary evaluations are in some form based on generalizations. Consequently, the issue of whether the evidence is sufficiently individualized must be assessed from case to case. The main premise is that suspicion cannot be solely based on group belonging. This is a legal-political requirement. Legal-political aspects consequently must also be taken into consideration when evaluating evidence.

A requirement of robustness

Even in cases where the evidence is evaluated so that a preponderance of the evidence exists and is sufficiently individualized, the evidence still needs to be sufficiently robust. Robustness is a measure of whether good grounds exist for the conclusion. It is not a question of how ”strong” the evidence is, but rather how thoroughly the situation has been investigated. Somewhat simplified, it is a question of ”how much” evidence exists rather than the strength of the individual pieces of evidence. This requirement of robustness varies depending upon time in that it is very low at the commencement of an investigation, but as time goes by, higher requirements are placed as to the ”completeness” of the investigation.