The Australian Financial Review

Letters, July 13, 2004


More Science, Less Hot Air

Alan Greenfield (Letters, July 8) needs to choose his words with more care.

Climate change is not “upon us” as he asserts. Rather, climate change is a natural phenomenon that has always been with us, and always will be.

Second, science does not operate on consensus, but on the basis of observation, experiment and tested theory.

Third, it is true that on some measures global surface temperature increased, in two steps, by a little less than a degree during the 20th century. The nature and magnitude of this change is consistent with its being related to one or more natural climatic cycles.

Finally, the degree to which the human activities may have contributed to the recent temperature rise remains extremely controversial among active climate researchers, though – judging from a recent report – apparently not among non-government organizations.

Understanding the complex climate-change issue, and explaining it to the public, requires a lot more science and a lot less propaganda.

Bob Carter
Marine Geophysical Laboratory

James Cook University
Townsville, Qld.


Michael Green replies (Letters, later)

Sad Reflection on the Nature of Science

Academic Bob Carter says “science does not operate by consensus, but on the basis of observation, experiment and tested theory” (“More science, less hot air”, July 13).

That a professional scientist feels confident to trot out this schoolboy methodology is a sad reflection of the unsophisticated understanding many scientists have about the nature of science.

First, peer review is a form of consensus.

Second, deciding what observations are important, what experiments are most relevant and what theories are sufficiently established are all issues involving scientific consensus. As is what should be in science textbooks and curricula.

Finally, climate change is not a laboratory science, in which a scientific consensus about the validity of a new hypothesis may emerge quickly from the results of a few carefully designed, repeatable experiments of isolated phenomena.

Climate change science involves complex interactions in a unique planetary environment. A huge range of actual and potential data, models, hypotheses, scientific disciplines and communities is involved. Much of the evidence is circumstantial.

It may be possible to establish the hypothesis that global warming trends are predominantly due to human activities on the balance of probabilities. But it may not be possible to establish it beyond reasonable doubt.

How established the hypothesis must be in order to support policy intervention and what interventions are appropriate and acceptable are not scientific questions.

Sure, our deliberations should be informed by high-quality science about which a wide consensus has been established.

But these issues are properly the subject of public debate and democratic political processes.

Michael Green
The Australian National University
Canberra, ACT


Bob Carter comments

I take solace from having been judged a professional scientist, and assert pride that the unsophisticated, schoolboy grasp that I have of scientific methods has sustained me so well through a 35-year-long career.

Many of Michael Green’s other points have little bearing on the matters raised in my original letter. They are mostly straw men, but nonetheless I address them briefly in turn below.

Peer review is not usefully described as a form of consensus. It is an editorial technique which serves as a convenient but not infallible quality control over scientific publication. Besides, it is not unusual for two or three independent referees (a “consensus”) to reject a paper for one journal only for different referees (another “consensus”) to recommend its publication by another. Quo vadis “consensus”?

Michael Green may decide what experiments to undertake by using a “consensus” of his peers, but most scientific breakthroughs come from mavericks who use their own brains to decide what is important experimentally and what is not. Neither is what should be in a science textbook decided by consensus; rather, it represents a series of professional judgements by the author(s).

Climate change is indeed a complex science, within which both experimentation and the collection of circumstantial evidence each play large roles. Science is not served - though politics may be - by introducing quasi-legal jargon such as the distinction between a “balance of probabilities” and proof “beyond reasonable doubt” to describe aspects of climate change. Such terminology has much to do with the law courts, and perhaps policy formulation, but very little to do with science.

Finally, of course public debate on such issues as climate change should be informed by high-quality science, which means that scientific hypotheses that are proposed as a basis for climate policy formulation must be exposed to rigorous probing and skeptical analysis. This is especially true when vested interests assert there to be a consensus on climate change, because the reality is that there is vigorous and healthy debate on the issue.

In keeping with this, western nations have spent more than $50 billion dollars on climate change research since 1990, to test the hypothesis that human-caused global warming is occurring. Despite this expenditure, and the huge efforts made by many talented scientists, no global anthropogenic climate signature, let alone a dangerous one, has yet been identified.


Bob Carter


Return to Comment & Opinion