I appologise for this being late. My computer and my ISP have been letting me down, the latter ithink due to a virus in my computer. Controllable Dose and the New ICRP Recommendations. In his early papers (Clarke 1998, Clarke 1999) Roger Clarke pointed out the lack of consistency between various recommended dose limits and action levels. The existence of different limits and different actions that depended on how the radiation exposure is received rather than the estimated risk to the person receiving it is confusing and illogical. The draft ICRP 2005 Recommendations only partly remove these problems and create further confusions. Dose limits. The problem with the present dose limits is that it is very hard to explain to the public why, with a natural background radiation of between 2 and 3 mSv per year in most places, but up to five times this in a few places, the public limit is an additional 1 mSv per year. That would seem to imply that some places in the world with high levels of natural radiation should not be inhabited. You can understand the confusion when you add the other limits: · about 6 mSv per year for household radon (not additive to any man made radiation exposure that the householder may receive), · 20 mSv per year for occupational exposures, but occasionally up to 50 mSv per year as long as the average is kept to 20. · 10 mSv per year and possibly up to 100 mSv per year for interventions. It would seem logical to recommend a dose limit that was applicable to all exposure situations. The word limit implies a level of risk that is the borderline of unacceptable risk. By having a single limit for all predictable exposures a logical system is created and it would be much easier to explain why there are different limits for the public, radon in homes and for radiation workers. Clarke suggested a set of limits starting at a maximum of 30 mSv per year and reducing by successive factors of ten to 30 microSv per year, This seemed a very reasonable starting point, although I would probably have used 20 mSv per year reducing to 300 microSv per year making 3 divisions rather than 4. He also suggested that the use of natural background as some sort of basis or comparison with regulatory limits would be useful in explaining limits to members of the general public. To make a limit of 30 microsievert per year when this is well within the variation in natural background and may be regularly received by frequent long distance flyers is risible. In fact to make a limit that is significantly less than natural background still leaves us with a problem in explanation. It implies that there is a known and measurable risk at this level of exposure, which is untrue. Dose constraints ICRP’s previous use of the word constraint was as a reduction factor of the limit, usually applied by a regulator, to reduce the exposure of individuals to a level that represented best available practice. This was a constraint on the practice partly as a form of ALARA and partly to take account of exposures that might be received from other practices. In the draft recommendations there are constraints that are larger than limits. This is going to be confusing if persisted with as it is a change from current interpretation of limit and constraint in relation to radiation exposure and it is in conflict with usual meaning of these words. While there are a number of dictionary definitions, the commonest ones are that a limit is a boundary that you must not go beyond and that a constraint is a restriction by a condition. These definitions seem to be exactly how the words have been interpreted in radiation protection in recent years. These meanings should continue and no constraint should be greater than a limit. Suggestions The draft recommendations should be closer to Clarke’s original proposals than they are. Specifically; · have one limit applicable to all radiation exposures, 20 mSv per year. · recommend several constraints for regulators to use, a factor of five down on the limit for occupational exposure, a factor of fifty down for public exposure including exposure from natural sources (in most parts of the world this would mean that the exposure due to practices would remain limited to about 1 mSv per year) and a factor of 100 down for exemption levels etc.. · recommend for exceptional circumstances a relaxation of a factor of 5 to cover accidents, interventions etc. This would result in a control system with limit + constraint values close to the present ones but with more logic and clarity.