What is the Kavli Institute for Theoretical Physics?

KITP is not a physics institute per se, but an institute for theoretical physics.

Asked to describe a scientist at work, people envision a white-lab-coat-clad person in a room filled with various apparatus.  That is the image of experiment, the carefully controlled probe or inquiry into the nature of Nature.

Experiment is one key to the modern scientific method, pioneered five centuries ago by Galileo. The other key is theory. Physicists specialize in either experiment/observation or theory.  Examples of theorists are Isaac Newton, Charles Darwin, and Albert Einstein.

The success of theory in physics dates from Newton, whose theory of mechanics and of gravity set the standard for theoretical physics for centuries---a concise mathematical framework that unified disparate phenomena, and had the power to make extremely precise predictions by means of which the theory could be tested and perhaps disproved.

Newton’s theory was an overwhelming confirmation of Galileo’s observation that “Mathematics is the language with which God has written the universe.”  The growing relevance of ever more advanced mathematics to theoretical inquiries has contributed to the emergence of theoretical physics as a specialty.

The ultimate product of science is the understanding of Nature in terms of successful theory.  Confusion about this role of theory in science occurs simply because in popular English language usage, the word “theory” is often taken to mean “conjecture” as “unproven idea.”   What is actually meant in terms of the role of theory in science—and the KITP mission--is the “idea proven,” with the degree of confidence quantified.

KITP programming is designed to maximize the probability of quality research outcomes through stimulation of interaction and collaboration among participants.

At KITP the in-house group consists of a handful of excellent physicists--permanent members--whose research interests are typically very different from one another, ranging from astrophysics, to particle physics, to string theory, to hard-condensed matter, to biophysics.  Also in-house are a distinguished group of postdoctoral fellows (newly degreed scientists with exceptional promise as researchers) whose interests are as diverse as those of the permanent members.  That in-house cadre of physicists provides guidance for the process whereby programs are formulated, selected and run.

The distinctiveness of KITP's model lies with the process whereby the programming is devised.   The process is characterized by a twofold purpose: (1) to encourage the broadest possible input for program proposals from the whole of the physics research community in order to generate a superb list of competing ideas; and then (2) to constitute a selection process among the competing ideas that enables the best possible programming.

Central to the process of program design is the role of the KITP Advisory Board, a group of distinguished scientists whose research interests (like those of the permanent members) differ widely and are therefore broadly representative of the physics community.  Members serve three-year terms so that the process of rotating membership ensures a steady stream of new perspectives on KITP programming choices.

Each year some 6,000 scientists, mostly physicists, are solicited to make suggestions for upcoming KITP programming for the upcoming year or two.  Members of the Advisory Board scrutinize the suggestions and winnow their number to the most promising, which may in fact represent a reformulation by an Advisory Board member or the combination of two or more suggestions.

Once a year the Advisory Board meets at KITP to engage in an arduous review process of the most promising proposals from which programming choices are then made for the next one to two years.  Along with choice of programs is the choice of program organizers—the scientists who take responsibility for running the program, devising its component sessions, and soliciting and selecting the best possible complement of participants.

In addition, one of the principal roles of the KITP Director is to solicit ideas that might not otherwise be submitted for programming from new communities that have yet to be represented at the KITP. A list of all KITP programs and conferences is located here. 

Proposals for programs must be based on a clear statement of the scientific problem, as well as a clear statement of why a program should run at a given time in the near future—i.e., launch of a new telescope requiring theoretical advances for the instrument’s best utilization or response of the theoretical community to an experimental discovery or a breakthrough in string theory needing further elaboration.

Selection of programming proposals depends on whether the science they entail is not only timely, but also exciting.  In addition, program organizers must be recognized leaders in their fields.  Good organizers attract good participants.

Organizers are asked to provide data describing the potential universe of program users.  That task entails organizers contacting up to 100 possible participants to determine whether sufficient interest exists to guarantee good participation throughout the proposed program duration.

One important aspect of the KITP modus operandi is the extensive preparation made by program organizers to ensure sustained quality among participants.  Programs are then announced one year to two years before they are to occur.  Would-be participants make application.  The usual selection rate is between 50 and 70 percent of applicants, with the more selective programs admitting one in three applicants.

The breadth of science done at KITP is remarkable. Ranging from cosmology to biology, from string theory to geophysics, to the more traditional physics fields such as condensed matter, particle physics, atomic physics, optics, turbulence, and complexity. To accomplish this requires a certain size and critical infrastructure, that precludes other similar, but more narrowly focused institutes from aspiring to such breadth of science.

The ability to encompass programming in so many areas of science often leads to synergies or interdisciplinary connections that simply cannot be done elsewhere.

Another related but special feature is the interdisciplinary nature of KITP research undertakings. Efforts include bringing physicists together with mathematicians or chemists or biologists.

For example, in the 1980s particle physicists and astrophysicists worked together at the KITP on the cosmic microwave background radiation in collaborations that helped engender the new field of particle astrophysics.

Recently, two programs running in parallel enabled atomic physicists, working on cold atoms, and condensed matter physicists, interested in strongly correlated fermionic systems, to create a whole new enterprise using techniques from the former to model the latter. That fused field is now one of the most dynamic areas of research in physics.

KITP's breadth of scientific programming provides a platform for innovation—facilitating conjunctions that lead to new scientific fields such as quantum computing (with the pioneering 1995 program) or new outreach endeavors such as programs for (1) secondary school teachers of science, (2) physicists at predominantly teaching institutions of higher learning, (3) physics graduate students, (4) professional communicators of science, (5) and an artist-in-residence.

Innovative activities are cost-effective at KITP because they arise from a breadth in scientific programming, as well as an established institutional infrastructure for hosting visitors.

The "norm" for scientific conferences (a widespread activity in the academic world) is to hold a series of talks illustrated via presentation software.  These talks are generally “set” pieces, given with incremental modifications at several conferences and guest seminar appearances. These set pieces are designed to take up 55 minutes of the hour customary allotted to speakers with five minutes at the end reserved for audience participation in the form of questions or comments that can only invoke short responses.

This traditional conference-talk model functions more as an advertisement for the speaker’s research, than as a collaborative exploration of ideas.  Traditional academic conferences may well be productive, however the give-and-take conversations about ideas happen not so much in the lecture hall, but outside in the hallways.

KITP's model is designed to break that mold by inducing speakers and audience members to interact more frequently and more authentically. The idea is to adapt the dynamic hallway-mode (typically involving two or three conversers) to the otherwise static lecture hall format.  It takes, of course, much more time to work together to explore ideas than to give or listen to traditional conference talks. KITP programs can afford to extend these dialogues, as they are measured in weeks and months, instead of days.

Formerly, conferences with their fixed lecture format served as the mode for the pre-publication dissemination of ideas and results.  Now, scientists post their papers on the Internet.

It also used to be (and still is, in many cases) that scientists would go to a conference, give talks, write up the presentations, and submit them to organizers responsible for producing a volume of conference proceedings appearing a year later when interest in the contents had likely declined.

KITP does hold 4-5 day conferences in conjunction with most programs.  Talks are available, generally within 24 hours, on-line via the KITP Online Talks Archive.  That archive consists of more than 10 years of some 10,000 talks.  The effort began with recording and posting KITP conference proceedings and later expanded to include talks in the programs.

Not every scientist can afford prolonged periods of time away from a home institution.  Conferences serve the population unable to commit to longer-term visits. This includes experimentalists whose research necessitates presence in their laboratories or observatories, but also theorists who for one reason or another cannot afford long leaves from home.

Conferences typically attract more participants than do programs, whose proceedings, moreover, are available online to brief those attending the shorter conferences.

Every effort is made to make the conference experience similar to the program experience in terms of encouraging interaction and collaboration.  Speakers are instructed to talk less and to take questions along the way from audience members, instead of deferring questions to the end.

In conjunction with most programs, the key purpose of conferences is to bring theory face to face (so to speak) with experiments or, in the case of astrophysics, observations.

Conferences held in conjunction with programs is a cost-effective use of KITP's infrastructure.  KITP conferences cost one-third of what it would cost to organize the same meetings as stand-alone events.

Research in conjunction with KITP programming has a greater impact on researchers than research conducted in conjunction with any other facility, according to a study whose findings were published in the Proceedings of the National Academy of Sciences. 

In addition to its overall effectiveness at promoting high impact research, KITP has helped create new fields.  Here are four examples:

• In the early 1980s, a group of young astrophysicists and particle physicists met for six months at the KITP and created the basic theoretical structure that governed a 25-year effort to understand the cosmic microwave background radiation.  Many of the participants became leaders in the new field of particle astrophysics that emerged from that KITP program.

They pioneered the theoretical modeling tools that led to the ability to use cosmic microwave background temperature measurements to construct an almost complete history of the universe and eventually to select the model incorporating dark matter and dark energy.

• From the 1995 program on quantum computing came the proof-of-concept that a quantum computer would be more powerful than classical binary computing systems—envisioning, in effect, the software that would make the quest for hardware not only a worthwhile, but potentially transformative technology.

• In 1998 the KITP hosted the annual string theory conference in conjunction with a six-month string-theory program.  String theory envisions all particles detected or hypothesized in high-energy physics as different vibrations of strings in five or six more “extra” spatial dimensions than the traditionally posited three.

The KITP string meeting occurred a month after Juan Maldacena announced his breakthrough insight into a duality that provided theorists with extraordinary computational power.  KITP permanent member and leading string theorist Jospeh Polchinski did a study on the effectiveness of that string theory program, indicating that the program had an enormous impact on the field—bringing all the top string theorists together at precisely the right time.

• In 2004 two programs ran simultaneously—one on cold atom systems or Bose-Einstein condensates and the other on strongly correlated electron systems. Bose-Einstein condensation (a big development in atomic physics acknowledged with more than one Nobel Prize) produced a coherent quantum state of cold atoms that Einstein had predicted. The big realization, given impetus via the conjunction of the two programs, is that these coherent quantum states can be used to construct models of condensed matter systems that are more controllable than had been imagined and attained heretofore. 

These two communities—one focusing on cold atoms (or, more broadly speaking, quantum optics) and the other focusing on highly correlated electron systems—had little, if any connection, until this concatenation at KITP that is now at the heart of the new field emerging from their fusion. The programs were not accidentally run together because it looked as though the two might benefit from a closer acquaintance, but the magnitude of the actual benefit was unanticipated.

Then there are the pedagogical programs that provide intense cross-disciplinary immersions—of biologists into physics, and physicists into biology, for one example. Another example is the cross between physics and mathematics. These pedagogical programs have emerged as the most frequently accessed of talks in KITP Online Talks. These instructive talks continue a service to the science community.

KITP research programs are designed by the best scientists.  KITP's Advisory Board, in conjunction with the director and permanent members, act as a conduit for input on programming solicited from the physics community.  Those topnotch scientists on the Advisory Board and in-house meet and discuss the submitted program proposals, and then select and help to shape the programs.  Quality programming emerges from this formative process.

Scientists design research programs for scientists, and doing so determines how money allocated for research at KITP is spent year by year.  Active successful researchers make decisions about the research programs to support.  The funded research programs succeed in a highly competitive winnowing that occurs in the annual process of KITP program creation.  Decision-making is kept closely connected to science.

KITP provides a highly flexible, cost-effective and adaptive platform for doing science.  That platform is designed to run the best programs, geared to ongoing basic science research that is very wide-ranging.  Long programs comprise the core of the KITP model because deep real-time human interactions—collaboration--take time, and are highly stimulating and productive of breakthroughs and the high impact research for which KITP has been recognized as number one. In addition to and in conjunction with programs, there are shorter conferences which draw a wider audience.