Making a statement about the age of the Universe implies the fundamental assumption that the Universe had a beginning.
Although we can easily determine that most constituents of our visible Universe, such as the islands of Hawai‘i, our Sun, the Earth, the Moon and the stars that we see at night, had a beginning, it is much more difficult to determine if the ensemble of all those things had a starting point in the past. One could imagine that the Universe itself has always been there, but that all of its components––the atoms, the planets and the stars––go through a continuous cycle of birth, life and death.
Until recently, this view of an eternal universe has dominated the history of philosophy. Today, we talk much about the Big Bang theory, which is a very solid and consistent physical theory of the origin and the evolution of the universe as we now observe it. We forget that for most of human history, an eternal universe was accepted as the most economical way of explaining the Cosmos.
Even today there is an independent physical theory of the Universe called the "Steady State Universe." This theory, supported by a small minority of astronomers, defends a model of an eternal universe where the well-known expansion of the Universe is countered by the continuous creation of one hydrogen atom per 10,000 cubic meters per year (this is a very small amount of matter) to maintain its average density constant. Hence the Universe should appear the same in all directions (this is isotropy) and at all distances and epochs (this is homogeneity).
The idea that the Universe had a beginning is relatively new and quite revolutionary.
Drilling into the History of the Universe
Admitting a beginning is philosophically awkward, and the nagging question of what was before naturally arises. What the modern physicist or astronomer has to accept is that the concept of a beginning, such as the Big Bang, also acknowledges the impossibility of looking further back in time than this point using the laws of physics. The intellectual positions of present day scientists and philosophers are more modest than that of Dr. John Lightfoot (1601-1675), Vice-Chancellor of the University of Cambridge, who declared that heaven, Earth and man were created on October 23, 4004 BC, at nine o’clock in the morning!
Did our Universe have a beginning? If so, how old is it? How do we determine its age? These questions are among the most difficult and also the most exciting addressed by modern astrophysics. The question of the Universe's age is a basic driver of contemporary cosmology. This quest has motivated the funding of many sophisticated ground-based and space experiments and is mobilizing the efforts of hundreds of scientists and engineers.
The cosmological questions are not new, but several millennia of exploration and the modern tools of physics and astronomy allow us a new perspective, and hopefully provide some credible answers. We can now give outstandingly precise answers to questions about the age of many of the Universe's components, such as the ages of cosmic bodies like the Earth and the Sun and their chemical constituents. It follows that the Universe itself should be at least older that any of its oldest components.
Light travels at about 300,000 kilometers per second in the vacuum of space. The powerful telescopes on Mauna Kea allow us to observe galaxies at distances of billions of light-years. This means that we capture particles (photons) of light that were emitted long ago and have traveled through expanding space eventually hitting the detectors mounted on our telescopes. These momentous tiny collisions end an incredibly long journey, but give us important clues on the Universe afar.
Thus, looking at distant galaxies is like drilling into the past and seeing the ancient history of the Universe.
A critical question that astronomers ask when determining the age of the Universe is how nearby galaxies compare to far-off galaxies, which we are seeing as they were at a time closer to their suspected birth. If the Universe began 13.7 billion years ago, as the latest data suggest, should galaxies at 6, 8 or even 10 billion light-years away look different? Theoretically, the answer is yes. Hubble Space Telescope images of the distant Universe tend to support this view that galaxies that are half or a third of the presently accepted age of the universe are different. However, we know that the collection of objects looked at by Hubble is limited.
What is the Age of Things?
Even the fundamental building blocks of matter like the proton and the electron are not eternal. We have known since the early 1970s that protons have colossal lifetimes, on the order of 1031 or 1032 years. We can state for sure that the constituents of the Universe are at least younger than 1031 years, which is a stunningly large number, but not a very useful conclusion.
A simple, but more robust way to infer an approximate age of the Universe is to calculate the time it takes for large structures, like galaxies and clusters of galaxies, to stabilize due to gravitational attraction. Just as flying sand particles slowly come to rest after a strong wind, disturbed systems in space take a period of time to come to a state of rest. Straightforward physics shows that under the attraction of gravity, primordial clouds of gas, star systems and galaxy clusters collapse to form well-defined structures that can be understood and over timescales that can be calculated.
The laws of physics also tend to constrain cosmic objects into flattened features (like the solar system, or the disks of spiral galaxies) or highly symmetrical structures (like clusters of stars and of galaxies). For example, we see millions of stable, well-settled galaxies throughout the nearby Universe. We can infer from their regular spiral or football shapes that their ages are much greater than 100 million years. Therefore, we know that the Universe must be much older than this.
On the other hand, we observe many clusters of galaxies that are still in the process of assembly. It is calculated that it can take as long as 40 billion years for these large systems to stabilize, which suggests that the Universe is not yet old enough for them to have reached equilibrium. Hence, one can conclude that the age of the Universe must be less than 40 billion years.
Radioactive Elements and the Evolving Colours of Stars as Cosmic Clocks
Although useful, the arguments above give us just a gross approximation of the possible ages of the Universe and of its constituents. We need better tools.
The disintegration of radioactive elements is one of the most powerful and accurate ways to derive the ages of astronomical systems. Like we use radiocarbon dating to infer the age of recent archaeological artefacts of 50,000 years or less, geophysicists and astronomers have used the properties of uranium isotopes and other radioactive elements to deduce an age of about 4.5 billion years for the Earth.
We can even determine that uranium and most of the heavy elements themselves found on the Earth and the Moon were produced about 8.8 billion years ago. They were probably "cooked" in a powerful supernova that exploded somewhere in the Milky Way and polluted the primordial interstellar cloud of gas and dust that would later be used in the formation of our solar system. Hence, the Universe must be older than 8.8 billion years, the age of the supernova “mother star” that produced the heaviest elements found on Earth.
One other powerful technique can be used to infer the age of the Universe to a finer precision. It employs the tracking of the colour and the luminosity of stars as they evolve during their long lifetimes. Images of stellar clusters taken through several filters allow astronomers to display of the systematic patterns of luminosity versus colour that betray ages and other properties of the stars. Globular or open clusters each contain thousands and even million of stars giving instantaneous “portraits” for given ages. By looking at different clusters, astronomers can plot colour-magnitude diagrams for many clusters. Each cluster will give a different pattern depending on its age, whether it is only a few million or 10 billion years old. These changing patterns of star luminosity and colour can be well reproduced through sophisticated computer modeling of stellar evolution. Astronomers have used this approach for half a century to infer the age of stellar clusters and their stars.
The oldest ages, inferred from studies of star clusters, are between 12 and 15 billion years, with uncertainties of a few billions years. Nevertheless these numbers tell us something reasonably certain about the oldest stars––they have to be younger than the age of the Universe. However, the venerable age of the oldest star clusters leaves an uncomfortably short amount of time between the current estimates for the birth of the Universe at about 13.7 billion years and the formation of the first stars.
Running the Movie of the Universe Backward
All of the previous examples indicate to us that the constituents of the Universe have a finite age. However that does not preclude the fact that the Universe could be eternal. The strongest evidence that the whole Universe had a beginning is the fact that it is expanding, which was discovered by Edwin Hubble (1889-1953) and Milton Humason (1891-1972) in the 1930s.
The expansion of the Universe puts tight limitations on everything. The fact that on a scale greater than a few million light-years all galaxies seem to be flying apart, surfing in expanding space-time, indicating that all things were together at some time in the past. The expansion of the Universe, once we accurately know the density of matter and energy, can be turned around to infer a very accurate birthday for the whole process. Like running a movie backward, astrophysicists can crunch everything existing in the present Universe into some sort of incredibly small and hyperdense nugget of space-time 13.7 billion years ago at the instant of the Big Bang––the time when everything, including space and time, began.
This prediction has implications on how things should look as our telescopes pierce further and further into the Universe. For example, when we observe distant galaxies, shining as they were 8 to 10 billions years ago, these “infant” galaxies should look different, very different, from the nearby more “mature” galaxies we see today.
The Gemini Deep Deep Survey of the Distant Universe
The recently completed Gemini Deep Deep Survey took the deepest spectra ever of very distant galaxies. Using data obtained with the Frederick C. Gillett Gemini North Telescope on Mauna Kea, a team of USA/Canada/UK astronomers have completed the analysis of the images and spectra representing several hundreds of galaxies corresponding to a time window when the Universe was between a 20% to 40% of its present age. They find a very puzzling “landscape.” The galaxy populations they encounter look the same, with surprisingly no sign of evolution during this crucial epoch that was thought to be a period of most significant changes in the assembly of galaxies. Even more intriguing: massive and fully formed galaxies are found at the largest distances, or youngest epochs of the Gemini survey. The big massive ones should not be there, but they are.
This finding leaves very little time between the Big Bang and that epoch for forming these Gemini Deep Deep Survey galaxies. Either something is wrong with our present models of collapsing large structures right after the Big Bang, or we need to revisit the way galaxies formed. For instance, massive black holes could be much more ubiquitous than we thought in the early Universe and may act as numerous and efficient seeds to form the first galaxies.
Although many indicators lead to an age of about 13 billions years for the universe and its constituents, there is an uncomfortably short period between the beginning of space-time-matter and the appearance of the first objects in the known Universe.