Sir Harold Jeffreys: Pioneer of Bayesian Probability and Geophysics
Sir Harold Jeffreys (1891–1989) was a towering figure in British science, whose intellectual reach spanned the rigorous domains of mathematics, statistics, and geophysics. A lifelong fellow of St John's College, Cambridge, Jeffreys is perhaps best remembered for his pivotal role in reviving the objective Bayesian view of probability, a framework that continues to influence modern data analysis and scientific inference.
Throughout a career that lasted over seven decades, Jeffreys balanced theoretical mathematical physics with the practical study of the Earth's physical constitution. His work provided essential tools for understanding everything from the quantum behavior of particles to the seismic vibrations of the planet.
Key Facts
- Primary Fields: Mathematics, Statistics, and Geophysics.
- Major Contribution: Revived objective Bayesian probability through his 1939 book, Theory of Probability.
- Academic Tenure: Fellow of St John's College, Cambridge, from 1914 until his death in 1989.
- Key Theory: Proposed that the Earth's planetary core was liquid.
- Notable Paradox: Associated with the Jeffreys–Lindley paradox.
- Scientific Stance: Remained a lifelong opponent of continental drift and plate tectonics.
Early Life and Academic Excellence
Born on April 22, 1891, in Fatfield, County Durham, Harold Jeffreys was the son of a headmaster and a school teacher. His early education took place at his father's school and Rutherford Technical College, followed by studies at Armstrong College in Newcastle upon Tyne and the University of London External Programme.
Jeffreys' academic prowess became evident at St John's College, Cambridge, where he studied the Mathematical Tripos. He earned first-class marks in Part One, became a Wrangler in Part Two, and received the prestigious Smith's Prize in 1915. These achievements paved the way for his lifelong association with Cambridge, where he eventually served as the Plumian Professor of Astronomy.
Contributions to Mathematics and Statistics
Jeffreys made profound impacts on how scientists interpret data. His work on Bayesian probability—a method of statistical inference in which Bayes' theorem is used to update the probability for a hypothesis as more evidence becomes available—was central to his legacy. He introduced the Jeffreys prior, a non-informative prior distribution used in Bayesian inference.
In the realm of mathematical physics, Jeffreys developed a general method for approximating solutions to linear, second-order differential equations by 1924. This work included applications to the Schrödinger equation (the fundamental equation of quantum mechanics). Because this work preceded the similar findings of Wentzel, Kramers, and Brillouin, the method is sometimes referred to as the WKBJ approximation to acknowledge Jeffreys' priority.
Collaborations and Publications
Jeffreys' intellectual partnership extended to his personal life; he married fellow mathematician and physicist Bertha Swirles in 1940. Together, they co-authored the influential text Methods of Mathematical Physics. His other seminal works, such as Scientific Inference and Theory of Probability, established him as a leader in the logic of science.
Geophysics and the Earth's Constitution
Beyond pure mathematics, Jeffreys was a distinguished geophysicist. He served as the Director of the International Seismological Summary (now the International Seismological Centre) from 1939 to 1952. One of his most significant geophysical assertions was that the Earth's core was liquid, a claim supported by his analysis of seismic data.
However, Jeffreys is also known for his steadfast opposition to the theory of continental drift proposed by Alfred Wegener and Arthur Holmes. He argued that there was no known physical force strong enough to move continents across the Earth's surface. Despite the mounting evidence for plate tectonics in the 1960s, Jeffreys remained a critic of the theory until his death in 1989.
Summary of Career and Honors
| Category | Details |
|---|---|
| Education | Armstrong College, University of London, St John's College (Cambridge) |
| Key Roles | Plumian Professor of Astronomy; Director of International Seismological Summary |
| Major Awards | Copley Medal, Royal Medal, Vetlesen Prize, Wollaston Medal, Guy Medal (Gold) |
| Notable Students | Hermann Bondi, Sydney Goldstein, Vasant Huzurbazar |
| Knighted | 1953 |
Frequently Asked Questions
What is the Jeffreys prior?
The Jeffreys prior is a non-informative prior distribution used in Bayesian statistics. It is designed to be invariant under reparameterization, meaning the results of the statistical inference remain consistent regardless of how the parameters are scaled or transformed.
Why did Sir Harold Jeffreys oppose continental drift?
Jeffreys believed that the physical forces required to move massive continents across the Earth's surface were non-existent or insufficient. He maintained this stance based on the geophysical evidence available to him, even as plate tectonics became the accepted scientific consensus.
What is the WKBJ approximation?
The WKBJ approximation is a method for finding approximate solutions to linear second-order differential equations. While often called the WKB approximation, the 'J' stands for Jeffreys, who developed the method in 1924, prior to the work of Wentzel, Kramers, and Brillouin.
How did Jeffreys influence the study of probability?
Through his book Theory of Probability (1939), Jeffreys helped revive the objective Bayesian approach. He argued that probability could be treated as a tool for logical inference based on available evidence, rather than just a measure of subjective belief or frequency.