In public arguments about controversial topics, one often hears the somewhat pugnaciously offered challenge: "Do you have a reference for that?" Meaning: Do you have a peer-reviewed confirmation of what you are saying, published in a reputable scientific journal? Providing such a confirmation usually lowers the temperature of the discussion somewhat.
Now, factual statements should certainly be backed by convincing factual evidence – that is the whole point of the empirical scientific approach. But apart from ignoring the difficulty that factual evidence means something rather different in, say, particle physics than in psychology, simplistic efforts to "have science on one's side" degrade science in the public eye to a caricature of itself, a stodgy guardian of indisputable facts, a records clerk. It is worth reminding ourselves that science is no such thing.
Far from being a boring fact-monger, scientific inquiry owes a profound debt to imagination. After all, the very idea of investigation is an exercise in speculative thinking, since someone must first wonder what is there to see in the depths of the sky before a telescope is actually trained at the stars. And beyond that indispensable moment of "What if?" which goes before any investigation, significant and far-reaching discoveries have often been a result of thinking in an adventurous direction. Let me recall a few from the history of my favorite science, physics.
In the face of the failure of all known physics to account for the radiation spectrum of a glowing furnace (the proverbial "black body"), Max Planck made the seemingly wild conjecture that perhaps the radiated energy comes in discrete packets, quanta. Planck had no "reference" for that, and nothing in the observed radiation spectrum cried out for that particular conjecture. But when he applied it, it described the spectrum perfectly, and of course, Planck's conjecture went on to be the central insight of all of quantum mechanics.
Similarly, Albert Einstein grappled with the thorny problem of reconciling the venerable discipline of Newtonian mechanics with the younger but very convincing theory of the electromagnetic field. These two theories differ in their mathematical structure and physical implications, and they could not both be correct. Einstein is said to have contemplated at length what it would be like to ride on a beam of light (an electromagnetic wave), as one would ride in a plane or a spaceship. To make the story short, it turns out that we, being massive objects, could never attain that speed. Moreover, on the beam of light there are no distances in time: to the light everything happens at once. The intuitive and well-established mechanics of Newton proved to be unsuitable close to light speed, and Einstein's improved description of the world is now known as special relativity.
Another notable contemplation of Einstein's was, "Why do things have only one mass?" The mass of an object manifests itself in two ways: as inertia in free motion, and as weight, the gravitational attraction. On the face of it, these two phenomena have little in common, yet the forces, accelerations etc. all speak of one and the same mass. Why should that be? Why are things that weigh heavy on the scale also sluggish to speed up? This counter-intuitive question led to the insight that gravitational pull is "free motion" in the space that is itself bent by gravity, the core insight of general relativity.
(Many introductions to the theories of relativity have been written; one of the more approachable and lucid ones can be found in this charming book.)
Vortices in thin air
There is, however, a difficulty with brilliant imagination: it does not in itself mean that you are right. In the waning days of pre-modern physics, the prevalent opinion held that the light, the electromagnetic wave, propagates by means of an invisible, tenuous, frictionless, non-viscous fluid called luminiferous aether. Every other kind of wave was known to need some material stuff to propagate through, so the aether hypothesis was a plausible but not very insightful reasoning by analogy – with one possible exception.
Vortices in an ideal fluid are eternal, and their lines of swirl cannot be broken. A "smoke ring" in aether remains a ring forever; a knotted vortex loop can never be untied, and so on. Physicist William Thomson (Lord Kelvin) hypothesized that there might exist primordial vortices in the all-pervading sea of aether: there would be various types of them, differing in their properties according to the topology of the knots they formed, eternal and immutable. They would be the atoms of matter.
In reality, ever more extravagant properties had to be attributed to the luminiferous aether in order to explain how it could carry the electromagnetic wave, and the aether finally evaporated in the historic Michelson-Morley experiment. Thomson's hypothesis of atomic vortices evaporated with it, and in any case, the actual atom turned out to be something quite different from a knotted vortex. All the same, one cannot but admire certain daring cleverness of this idea, bringing together the subtleties of fluid mechanics and topology in an attempt to account for something seemingly unrelated: the existence and varied properties of atoms. It would have been remarkable had it turned out to be true.
Of course, the correct theory of the atom, the quantum mechanics, more than compensates for this "loss" with its own remarkable subtlety. But we can hear the echoes of the vortex theory in the speculative ideas of our own time: such as the string theory, which posits that the ultimate building blocks of the universe are not dot-like but line-like; or the idea of the space-time as a "quantum foam" teeming with the fluctuations of the quantum uncertainty principle. These are attempts to stretch our current understanding of physics beyond what we can observe at present, in order to explain things we still don't understand, chiefly how gravity and quantum mechanics fit together. Only future experiments will decide whether these conjectures describe something real, or are they merely clever flights of fancy, like Thomson's vortices.
Imagination in a straitjacket
This brings us to a remark by the physicist Richard Feynman, who once gave a memorably succinct description of the scientific process: he called it "imagination in a straitjacket." What Feynman meant is that, as scientists, we accept that all insights, however flighty or clever, must eventually be subjected to the straitjacket of the experiment, of empirical verification, for that is the great strength of science.
But on the other hand, we also know that imagination withers in a straitjacket: overly constrained insight loses courage and gradually gives way to low-risk squabbles over evidentiary minutiae. In contrast, it is the inevitable fate of any far-reaching understanding that it skates on thin ice of evidence, because the methods of observation have yet to catch up with it; the right straitjacket has yet to be devised. As another historical example, irrefutable evidence of evolution was not available to Charles Darwin in his lifetime, nor was molecular genetics, nor computational modeling of evolutionary processes; still, his elegant insight into the dynamics of living things has withstood the test of time.
So perhaps we should refine Feynman's dictum. The jacket is beneficial and necessary because it protects empirical truth from error, deliberate falsehoods and outright quackery. Moreover, when experiments that would answer crucial questions are beyond practical reach – when an adequate jacket cannot be devised, as is currently the case in much of the fundamental physics – extravagant speculative theories begin to proliferate freely, as they once did in the days of the luminiferous aether.
But the very accumulation of knowledge inexorably tightens the straitjacket: as a scientific field matures, progress becomes constrained by what is known, and runs the risk of becoming risk-averse and timid. In this light, we should uphold the importance of imagination and creativity, and defend their legitimate place in the ongoing dance of conjecture and verification that is the scientific enterprise. Not every age is conducive to mind's grand adventure, and not every daring idea is true. But should we all be content merely to safely reference each other's impeccable results? Without answering the seductive call of the unknown and the adventurous, where would new knowledge come from?