Our biggest concern is 53b.

"53b. There should be a wide definition of impacts, including economic, social, public policy, cultural and quality of life. ... (Within the `impact' element we do not intend to include impact through intellectual influence on scientific knowledge and academia - this is fully recognised within the `outputs' and `environment' elements of the REF.... "

This narrow definition of impact is a grave error. Economic benefits are an important by-product of academic research --- they are not the main reason for doing academic research; nor are they the fundamental driving factor[1]. The primary purpose of academic research is to discover the truth about our world. Any assessment of impact which fails to account for this, most important, aspect, will be fundamentally flawed.

Therefore, we propose that HEFCE broaden the definition of impact to include all important impacts of research: including social, economic, scientific, and intellectual. If this change is not made, then the consequences will be deeply damaging.

First, research that is primarily foundational and long-term in nature will be heavily penalised. Under the proposed scheme, it would be entirely possible that pure research of the highest academic standard would not be regarded as 4*, simply because it fails against the narrow and misguided notion of impact being proposed. Inevitably this will result in such research being devalued in the UK research base, with damaging long term consequences. Why should a leading researcher stay in the UK when the UK's research evaluation system is designed so that it cannot recognise their quality? There is much research of the very highest international standard in the UK research base that does not meet the narrow criteria being proposed, and which therefore could not be awarded the highest rank under the proposed scheme. This is just ludicrous.

Second, if the notion of impact is not changed, then it will not be possible to draw meaningful conclusions from the results of the REF. The problem is that credit is awarded for two very different characteristics that research might have ``this research is excellent (independently of whether it is fundamental research or applied research)" and ``this research is applied''. Since, according to the proposal, the measure of excellence will be conflated with the measure of whether or not research is applied, it will be impossible to interpret the results. If one department gets a higher score than another, that will either mean that the first department has better research, or else it will just mean that the first department happens to have more applied researchers. With the numbers mixed up, nobody will know which.

If HEFCE is unable to broaden the definition of impact so that it applies equally to all research, then we suggest that the name of the exercise should be changed to ``Research Excellence and Economic Impact Framework'' so that the public is not misled by the outcome.