Cosmic inflation has been at the centre of our story of the Big Bang for decades. But the theory has many problems, resulting in wildly speculative proposals like the "inflaton field" and some argue the theory breaks the light speed limit. Theoretical physicists Daniele Bertacca and Raul Jimenez here argue the cosmic inflation paradigm has too many free parameters, and we are expanding the limits of our theories to fit the data. Bertacca and Jimenez propose a radical alternative: gravitational waves and their dynamics were key to the early universe and ultimately gave rise to galaxies, stars, and planets.

This article is presented in association with Closer to Truth, an award-winning broadcast and digital media series that has run continuously since 2000. Closer to Truth's mission is to explore humanity's deepest questions with the world's leading thinkers and scientists. Closer to Truth is also a partner of the upcoming HowTheLightGetsIn festival in London this September 20-21, where you'll find debates on cutting-edge cosmology, science, philosophy, and politics, with world-leading thinkers like Sir Roger Penrose, Sabine Hossenfelder, John Gray, and many more. 

One, if not the, guiding principle in physics is that of the Ockham razor: the simplest model with the least number of parameters is the one that describes nature. A stronger view is that nature should not be described by models but by what we term “rigid theories,” a group of equations or concepts that do not contain any free parameters.

In physics, whenever we have found a theory that describes nature, it has been because of its “rigidity.” Newton's law of universal gravitation establishes that the force between two masses is inversely proportional to the square of the distance between two objects. This describes most data in the sky, and in a laboratory on Earth regarding the gravitational attraction between two masses. But why the square of the distance and not the cube of it, or 3.5 times it? The orbits of celestial objects would be different, but this would be purely an observational fact.

___

In science, too much flexibility can be problematic, as it makes it difficult to know whether a model is actually predicting something or simply fitting data after the fact.

___

Indeed, Newton's law is just, a very clever, fit to Kepler's data. It is not surprising then that the new tools in the field of computer science like neural networks and large language models can easily find Newton’s law from observations of the orbits of celestial objects; after all they are the universal interpolator in the body of (scanned) human knowledge.