Home > Positioning > Persons > Carroll

Sean Carroll (1966–)

Carroll is a theoretical physicist and cosmologist whose work addresses the question of why the universe has an arrow of time — why entropy increases, why we remember the past and not the future, and what the low-entropy initial condition of the universe means for cosmology and for the foundations of physics. From Eternity to Here: The Quest for the Ultimate Theory of Time (2010) is the most sustained and accessible treatment of the arrow-of-time question since Eddington named it. Carroll has also been a prominent defender of the Everett (many-worlds) interpretation of quantum mechanics and a public voice for the view that physics and philosophy need each other — arguing against the “shut up and calculate” attitude that treats interpretive questions as unscientific.

Life

Born 5 October 1966 in Philadelphia, Pennsylvania. Undergraduate at Villanova University (BS, 1988). PhD in astronomy and astrophysics at Harvard (1993). Research associate at the MIT Center for Theoretical Physics, then at the University of Chicago. Senior research associate at Caltech (2006–20) — a research appointment, not a tenure-track position, which reflected both the interdisciplinary breadth of his work and the institutional structure of theoretical physics appointments. Homewood Professor of Natural Philosophy at Johns Hopkins University (2022–), an appointment that reflects his combined physics and philosophy practice.

Carroll has been unusually active as a public communicator — his blog (Preposterous Universe), his podcast (Mindscape), and his popular books have made him one of the most visible theoretical physicists in public discourse. He has argued explicitly that outreach is part of the scientific responsibility, not a distraction from it. His philosophical engagement — with David Albert on the arrow of time, with the Everettian community on quantum foundations, and more broadly with the philosophy-of-physics community — distinguishes him from most working physicists.

The arrow of time and the Past Hypothesis

From Eternity to Here (2010) organises the arrow-of-time question around the low-entropy initial condition. The argument builds on Boltzmann’s statistical mechanics, Penrose’s quantification of the improbability of the initial state, and Albert’s formulation of the Past Hypothesis:

The fundamental equations of physics are time-symmetric — they work equally well run backwards. The arrow of time arises not from the dynamics but from the boundary condition: the early universe was in a state of extraordinarily low entropy. Without this boundary condition, the statistical mechanics would predict equilibrium in both temporal directions — no arrow, no records, no memory. “The reason we remember the past and not the future, the reason effects always follow causes and never vice versa, is because of entropy.”

Carroll’s contribution is to take the Past Hypothesis seriously as a cosmological problem — not just as a postulate to be accepted (as Albert argues) but as something that needs a cosmological explanation. His proposal: the universe we observe may be a local fluctuation within a much larger multiverse, in which low-entropy regions arise spontaneously and inevitably through a process of eternal inflation. The arrow of time is real within our region but has no global direction — different regions may have arrows pointing in different directions, and the question “why does the universe have a low-entropy beginning?” is answered by “our region does; other regions may not.”

The baby-universe scenario (developed with Jennifer Chen, 2004) proposes that new universes can nucleate from high-entropy de Sitter space through quantum fluctuations — each new universe begins with a low-entropy state and develops its own arrow of time. The mechanism is speculative but has the virtue of explaining the low-entropy beginning without postulating it as a brute fact: the beginning is a consequence of the dynamics of the multiverse, not an arbitrary initial condition.

The many-worlds interpretation

Carroll is a prominent defender of the Everett interpretation — the proposal that the wave function of the universe is the fundamental reality, that it evolves unitarily (no collapse), and that what appears as “measurement” is the entanglement of the observer with the measured system, producing branching in the universal wave function. In Something Deeply Hidden: Quantum Worlds and the Emergence of Spacetime (2019), Carroll argues that many-worlds is the most parsimonious interpretation: it takes the formalism at face value, adds no collapse postulate and no hidden variables, and avoids the measurement problem by denying that measurements have special status.

The central challenge for many-worlds is the probability problem: if every outcome of every measurement occurs (in different branches), what does it mean to say that one outcome is more probable than another? Carroll defends the self-locating-uncertainty approach: before the branching occurs, the observer is uncertain about which branch they will find themselves in, and the Born rule gives the correct probabilities for self-location. Albert has challenged this approach; the debate is active and unresolved.

Carroll connects the many-worlds interpretation to his cosmological work: if the wave function is the fundamental reality, then the multiverse is not a speculation added on top of quantum mechanics but a consequence of taking the formalism seriously. The inflationary multiverse and the many-worlds multiverse may be aspects of the same picture.

Where Carroll stops

The baby-universe scenario and the multiverse explanation of the arrow of time are speculative — they invoke cosmological structures (eternal inflation, baby universes, de Sitter space nucleation) that are not directly observable and may not be testable. Carroll has acknowledged this explicitly and argued that testability is a spectrum, not a binary: a theory can be assessed by its explanatory power, its internal consistency, and its fertility even when direct experimental tests are unavailable. Whether this is a principled expansion of the scientific method or an erosion of the testability criterion is debated. Lee Smolin and Peter Woit have argued that the multiverse programme (in which Carroll’s work participates) has moved beyond the reach of empirical science; Carroll has responded that the criterion of direct testability, strictly applied, would exclude much of established physics (general relativity’s predictions about the early universe, for example).

The many-worlds interpretation, as Carroll presents it, resolves the measurement problem by positing that all outcomes occur. Albert’s objection — that the probability problem is not solved by self-locating uncertainty — has not been conclusively met. The deeper question is whether many-worlds is genuinely parsimonious (one universal wave function, no collapse, no hidden variables) or whether it merely trades one set of postulates for another (the branching structure, the preferred basis, the interpretation of probability). The parsimony argument depends on what is counted as a postulate; Carroll and Albert disagree about the accounting.

Carroll’s public engagement — books, blog, podcast — has made him one of the most influential physicists in popular discourse, but the gap between his public profile and his academic position (he was a research associate at Caltech for over a decade without tenure) reflects a tension in the physics community about whether philosophy-of-physics work counts as physics. The Johns Hopkins appointment as Professor of Natural Philosophy may signal that this tension is resolving, but the institutional question — whether the foundations of physics is a legitimate subfield of physics or a branch of philosophy — is not settled.

Key works

See also: Boltzmann · Albert · Penrose