Most criminal justice systems accord a central role to an offender's criminal history in the determination of sentence, although the importance of previous convictions varies across jurisdictions. In countries such as Canada, Australia, England and Wales, criminal record plays a critical but limited role in sentencing. Elsewhere, in several American states, criminal record is primary, eclipsing even the seriousness of the offence of conviction as a determinant of sentence severity.(1) Criminal history is relevant in jurisdictions with formal guideline systems, as well as countries that have traditionally eschewed such attempts to structure sentencing practices. Most penal codes around the world also contain repetitive offender provisions that prescribe higher penalties for recidivists, or which defer parole eligibility for the recidivist offender.(2) Statutory premiums for recidivists are not new--for example, the Indictable Offences Bill, introduced in England well over a century ago, provided a mandatory minimum sentence for repeat offenders.(3) Finally, it is worth noting that public attitudes regarding sentencing are also strongly influenced by the offender's criminal antecedents. Survey research in the U.S., Canada, and elsewhere suggests that when members of the public are asked to "sentence" offenders, support for incarceration increases as a direct function of the number of prior convictions.(4) Justifying the "Recidivist Premium" The exact impact of criminal record is determined, in theory, by the nature of the underlying sentencing purpose(s). Some desert theorists, including Andrew von Hirsch and Martin Wasik,(5) regard criminal history as being relevant to the sentencing decision, albeit in a limited way. They argue that our everyday moral judgements incorporate the notion of leniency towards people who transgress for the first time. According to this view, a lapse into criminality after a lifetime of lawabiding behavior should be responded to with a certain degree of tolerance. The result is a first offender discount. As Wasik and Von Hirsch note: "A first offender, after being confronted with censure or blame, is capable--as a reasoning human being presumed capable of ethical judgments--of reflecting on the morality of what he has done and of making an extra effort to show greater restraint. What we do, in granting the discount [to first offenders], is to show respect for this capacity--and thereby give the offender a so-called 'second chance.' With repetitions, however, the discount should begin to diminish and eventually disappear."(6) In practice, this justification is known as the progressive loss of mitigation.(7) However, there are other desert theorists (such as George Fletcher and Alexis Durham III) who see little role for criminal record in the determination of sanction(8). According to their perspective, there should be no discount for first offenders, and no increase in sentence severity as a function of the number of previous convictions. Neither version of desert theory distinguishes between offenders with lengthy, but varying criminal histories. For Wasik and Von Hirsch, such offenders can no longer claim that their current transgressions are at odds with their actions to that point; they have had their "second chance." For the other desert theorists, there is no role for criminal record to mitigate sentence even for first offenders. The utilitarian sentencing purposes use an offender's criminal record in a very different way. An incapacitation strategy would give primary weight to the offender's criminal record, on grounds that the actuarial probability of future offending depends largely on the number and nature of priors. Advocates of individual deterrence would take a re-conviction as evidence that a harsher sentence was now required as the previous sentence clearly failed to deter the offender. Some form of recidivist premium then is justified according to almost all sentencing theories. …