If you knew you were going to lose your job within the next five years, you might just go looking for another one today, not waiting to see exactly when you would be let go. But, if your job were very lucrative, and all you knew was that after five years your pay could be reduced each year, you might want to weigh that against the uncertain prospects of getting a new job that may or may not match your old one.

The second situation is roughly analogous to what we face with the coming peak in world oil production. We know the peak will occur–some say it already has–but we don’t know exactly when. We know that supplies will plateau or decline, but we can’t know how long the plateau will last (if there is one), nor how steep any decline might be. Alternative fuels are now being developed and competing for a future share of the energy market, but no particular fuel has emerged as the answer to oil decline and, perhaps just as important, to global warming concerns.

With respect to oil peak, each combination of dates and decline curves suggests a different response. An imminent peak followed by a sharp decline–the worst possible combination–would demand immediate emergency conservation measures and massive investment in renewable energy research and deployment. A late peak followed by a long plateau suggests that we have more time to think through our response and gradually work toward solutions. Something in-between may call for a rapid implementation of conservation programs and a quick decision to back certain oil replacements based on the best available knowledge, but without the benefit of a long trial period in the marketplace.

All three scenarios are on offer in the peak oil literature. Energy consultant Robert Hirsch’s study of decline curves from oil-producing countries around the world suggests sharply declining oil supplies after the peak. Combine that with predictions of an imminent peak, and you get the first scenario above. Daniel Yergin, the ever-optimistic president of Cambridge Energy Research Associates, predicts not a peak but an “undulating plateau” for several decades beginning sometime after 2030 or 2040.

Douglas Reynolds, a resource economist at the University of Alaska-Fairbanks, believes a peak will occur sometime before 2015, but that it will take the form of a long, gradual curve–more like the bottom of a saucer than the peak of a mountain. Reynolds believes political factors as much as geological limits will cause this type of peak.

The permutations of these scenarios are many, and the available evidence doesn’t tell us definitively which one to expect. Henry Groppe, an oil forecaster with a remarkable record, believes we are at peak, but will face at least a decade-long plateau in the production of liquid hydrocarbons as growing volumes of natural gas liquids and condensates make up for declining oil production. This scenario offers some hope that an immediate peak does not spell the end of civilization as we know it, though it certainly portends much hardship, especially for the poor. On the other hand, a peak which comes 25 years hence may not seem that threatening today. But, if it is followed by a steep decline in oil supplies, it means we had better start making and implementing plans now to get ready. For why this is so, read Robert Hirsch’s report on mitigating the effects of peak oil prepared for the U. S. Department of Energy. Hirsch says that in order to avoid huge economic dislocations, the world would need to begin a crash problem to identify and deploy alternative liquid fuels at least 20 years in advance of any peak.

As for Groppe, he has 90 percent of his equity assets in energy; he puts his money where his mouth is. But, such investment stances can also create reinforcing loops in the minds of those who are committed to a particular view. Naturally, investment managers invest based on their research findings. But afterwards they may look for evidence to reinforce those decisions and downplay evidence that contradicts their initial research.

Groppe may be immune to this. But, we should be careful to parse the thinking behind various ideas about the date of the peak and the shape of the decline curve. Do the predictions we cling to come from our deepest fears, our greatest hopes, our ideological predilections or possibly even our investment portfolios? Or, are we trying to see the world as it is, adjusting our thinking and actions to unfolding events rather than preconceived notions?

All of us like to see our predictions vindicated. But we should let neither a hatred of the current globalized, corporate-dominated economic system nor a unthinking devotion to the free-market creed that is its handmaiden guide us in evaluating remaining oil and other finite energy supplies. When the stakes are this high–namely, the future of human civilization–we ought to focus on careful observations, flexible thinking and, most of all, humility.

That doesn’t mean we shouldn’t prepare for the future based on what we know and believe. When facing great uncertainty and large possible consequences, thorough preparation is the wise course. In fact, being ready early ought to be considered a virtue rather than a failure to predict the future correctly. Under the circumstances the last thing we need is a swaggering certainty in our pronouncements. That’s the one thing of which I would have thought the world had enough.

(For my recent discussions of uncertainty, risk and probability, see Can a Wall Street Maverick Tell Us Something About Our Ecological Future? and What if Daniel Yergin is Wrong?)