There's a very interesting article in the Independent newspaper in England about the problem of physics becoming separated from independent, experimental proof of its postulations.
Dark energy is the latest and most daunting puzzle to confront cosmologists, adding to another mystery that has haunted them for decades: dark matter. Nearly 90 per cent of the mass of galaxies seems to be made of matter that is unknown and unseen. We know it must be there, for without its gravitational pull the galaxies would have disintegrated. Cosmologists in particular and physicists in general, are now faced with the stark reality that roughly 96 per cent of the universe cannot be explained with the theories at hand. All our efforts to understand the material world have illuminated only a tiny fraction of the cosmos.
And there are other mysteries. What is the origin of mass? What happened to the anti matter that should have been produced along with matter during the big bang? After almost a century of success at explaining our world using the twin pillars of modern physics – quantum mechanics and Einstein's general theory of relativity – physicists have reached a plateau.
The way forward will involve reconciling quantum mechanics with general relativity into a theory of quantum gravity. In situations where the two domains collide – where overwhelming gravity meets microscopic volumes, such as in black holes or in a big bang – the theories don't work well together. In fact, they fail miserably. One of the most ambitious attempts to bring them together is string theory, an edifice of incredible mathematical complexity. Its most ardent proponents hope that it will lead us not just to quantum gravity but to a theory of everything, allowing us to describe every aspect of the universe with a few simple equations.
But the theory's hoped-for denouement is nowhere in sight. Far from explaining our universe, string theory seems to predict the existence of 10,500 universes or more. Crucially, the theory is so far from being verified experimentally that it has become the poster child of what is wrong with physics today. Theory has lost touch with experiments – and, so, with reality.
The greatest advances in physics have come when theory has moved in near-lockstep with experiment. Sometimes the theory has come first and sometimes it's the other way around. It was an experiment in 1887 by Albert Michelson and Edward Morley – showing that the speed of light is not dependant on the motion of the observer – that influenced Einstein's 1905 formulation of the special theory of relativity. A decade later, Einstein produced the general theory of relativity, but it was only after experiments in 1919 verified a fascinating implication of general relativity – the bending of starlight by the sun's gravity – that the theory gained widespread acceptance. And throughout the early to mid-1900s, theorists and experimentalists jostled and outdid each other as they shaped quantum mechanics. An equally fruitful collaboration occurred in the 1960s and 1970s, when particle physicists theorised about the fundamental particles and forces that make up the material world, and experiments confirmed their startlingly accurate predictions. But this energetic interplay is now deadlocked. The discovery of dark energy and dark matter, along with the failure, so far, of experiments to find the Higgs boson (thought to give elementary particles their mass), has allowed theorists free rein. Ideas abound, adrift in a sea of speculation. Can the next generation of experiments help anchor the theories to reality?
There's more at the link. Words in italics are my emphasis.
I couldn't help but compare the author's point to the state of climate research. Remember the models predicting anthropogenic global warming? They weren't based on observations or experimentally proven, repeatable results. They were theoretical models, constructed mathematically, which became the bedrock of the 'science' of climatic prediction. It turned out to be pseudo-science, of course . . . none of the models, when fed actual temperature and weather records from the past, could accurately portray what turned out to be the real-world results of those figures a few years later. Not one of the models has demonstrated any accuracy under rigidly-controlled experimental conditions. It looks like the problem has spread to other areas of science as well.
Fortunately, according to the author, there's a good deal of experimentation going on to tackle the immensely intricate questions raised by 'theoretical physics'. He mentions several in the article, and more in a book he's written. It's encouraging to see that scientific rigor is still being applied.