Few-Body versus Many-Body Physics

There is no universal particle number at which few-body physics suddenly becomes many-body physics. Three resonantly interacting particles can exhibit subtle universal structure, while a much larger noninteracting system may admit a simple mode-by-mode solution. A chain of 24 spins is finite and microscopic, yet it is already a genuine many-body problem.

The useful classification is therefore operational. It answers three independent questions:

1. Problem structure: can the active constituents or relative coordinates be resolved individually, or must the problem be organized through modes, correlations, locality, and collective variables?
2. Physical scale: is the system microscopic, mesoscopic, or macroscopic for the observable and resolution under discussion?
3. Claim type: is the conclusion an exact statement about one finite system, evidence from a sequence of finite systems, or a statement about a specified infinite-system limit?

A trapped gas of $10^5$ atoms, for example, can be a finite many-body system that is macroscopic for its central density profile. It is still not itself a thermodynamic limit. Keeping these axes separate prevents particle count from doing conceptual work it cannot do.

Helpful background. From Few-Body to Many-Body Physics supplies the chapter map, and Why Many-Body Physics Is Different motivates the change of organizing principles. Neither is a hard prerequisite.

Three Independent Meanings of “Many”

The following labels are useful, but they are not mutually exclusive boxes.

Label	What it classifies	Typical organizing question	What the label does not imply
few-body problem	problem and method regime	Can all active constituents, channels, or relative coordinates be tracked explicitly?	weak interactions, separability, or easy computation
finite many-body problem	problem and method regime	Do statistics, locality, correlation functions, reduced observables, or collective variables organize the calculation?	macroscopic size, thermal equilibrium, interactions, or an infinite system
macroscopic finite system	physical scale relative to an observable	Are microscopic, boundary, and level-spacing effects negligible at the stated resolution?	that every observable has reached its bulk limit
thermodynamic-limit claim	mathematical claim about a sequence	Does a specified observable converge along a specified family of systems?	that one large sample proves the limiting statement

Thus “finite many-body” and “macroscopic finite” may both describe one sample. “Thermodynamic limit” instead describes a controlled limiting construction, not a specimen on a laboratory table.

An Operational Classification Rubric

Before assigning a label, record five pieces of information.

Question	What to state
What are the active degrees of freedom?	particles, sites, modes, spins, bands, impurities, or collective coordinates
What is the size variable?	particle number $N$, number of sites $L$, volume $V$, number of modes, flavor number, or another parameter
What organizes the target calculation?	individual coordinates and channels, or correlations, local observables, response functions, and effective fields
Which observable and resolution matter?	support of the observable, distance to a boundary, energy and time resolution, desired accuracy, and relevant correlation length
What kind of claim is being made?	one finite result, a comparison across sizes, a controlled extrapolation, or an infinite-system theorem

A compact scientific statement has the form:

For the family of systems ___, with active degrees of freedom ___ and size parameter ___, we study the observable ___ at resolution ___. The result is a ___ claim, supported by ___.

This sentence is deliberately more informative than “we studied a large-$N$ system.” It exposes whether $N$ is a particle count, a symmetry rank, a number of sites, or merely a computational parameter.

Few-Body Problems Resolve the Constituents

A few-body problem has a fixed, order-one number of active constituents, coordinates, or scattering channels that can be treated individually. Typical questions include:

• which bound states exist;
• how incoming channels scatter into outgoing channels;
• how exchange symmetry constrains a small number of particles;
• which relative-coordinate correlations characterize a state;
• how a controlled approximation changes a few-body spectrum.

For three particles with positions $\mathbf r_1$, $\mathbf r_2$, and $\mathbf r_3$, one commonly removes center-of-mass motion and works with two relative coordinates. The details can still be formidable: three identical bosons in three dimensions can exhibit Efimov physics when short-range forces have a scattering length whose magnitude is much larger than their interaction range. Strong correlation and universality therefore do not, by themselves, make a system many-body.

“Few” has no context-independent upper cutoff. Four-body scattering is normally few-body physics. A ten-particle calculation might also be treated as few-body if all active coordinates or channels remain explicitly controlled. In another setting, ten local spins may already be approached with many-body language because the relevant questions concern entanglement, operator spreading, or the scaling of a lattice family.

Few-body is therefore a statement about the active description and target question, not a synonym for small matrix or easy problem.

Finite Many-Body Problems Reorganize the Questions

A finite many-body problem contains many active particles, sites, or modes, but it remains a finite quantum system. What changes is the useful language. Instead of attempting to describe every amplitude of a generic wavefunction, one usually asks about:

• local or few-body observables;
• reduced density operators and correlation functions;
• conserved sectors and symmetry-resolved spectra;
• response functions and collective modes;
• entanglement structure;
• scaling with system size, time, or resolution.

For a spin-$1/2$ chain with $L$ sites, the unconstrained state space has dimension $2^L$. That single illustration is enough here: the exact counting rules, symmetry reductions, and computational implications belong to Scaling of Hilbert Space.

Neither interactions nor temperature define this regime. An ideal Fermi gas is many-body because antisymmetry, occupations, and collective thermodynamic observables involve many particles even though the Hamiltonian is diagonal in single-particle modes. Conversely, a two-level atom in a canonical ensemble is statistical-mechanical but not a many-body system.

A finite many-body Hamiltonian can also contain only two-body terms:

$$H = \sum_i h_i + \sum_{i<j} V_{ij}.$$

Here “two-body” classifies the support of each interaction term, not the number of bodies in the system. The same distinction applies to a two-body observable such as $\langle n_i n_j\rangle$: it probes two sites inside a many-body state.

Macroscopic Finite Is Observable-Dependent

Macroscopicity compares the system with the scales resolved by a particular measurement. Let $L_{\mathrm{sys}}$ be a characteristic system size, $d_{\partial}$ the distance from the observable to a boundary, $R$ an interaction range, and $\xi_O$ a correlation or response length relevant to the observable $O$. A local bulk observable may behave approximately as in an infinite medium when, schematically,

$$L_{\mathrm{sys}},\,d_{\partial} \gg R,\,\xi_O,$$

provided no long-range interaction, critical mode, topological edge contribution, or other mechanism invalidates that scale separation.

This test is observable-specific. In one open, gapped spin chain:

• the central energy density may already be insensitive to the ends;
• an edge polarization remains boundary-sensitive by definition;
• an exponentially small sector splitting can still depend strongly on $L$;
• the global gap and recurrence time remain properties of the finite spectrum;
• a long-range correlator may require a much larger system than a one-site observable.

The same sample can therefore be bulk-like for one measurement and visibly finite for another.

At regular positive temperature, a finite-dimensional Hamiltonian that depends analytically on its parameters has a finite partition function

$$Z_L(\beta,g) = \operatorname{Tr} e^{-\beta H_L(g)}$$

and an analytic free energy wherever these assumptions hold. A sharp peak or rapid crossover in one finite sample is not, by itself, a thermodynamic phase transition. This statement needs a zero-temperature caveat: exact level crossings, often protected by symmetry, can make a finite-system ground-state energy nonanalytic. The full relation between finite systems, singular limits, and noncommuting limits belongs to Thermodynamic Limit.

Why Particle Number Alone Fails

The same count can conceal very different physics.

Feature that matters	Why it changes the classification or method	Counterexample to a naive particle-count rule
active degrees of freedom	a nominal constituent may be frozen out, while an environment may supply many active excitations	one mobile impurity coupled to a Fermi sea is a many-body problem
exchange statistics	identical-particle sectors and occupations reorganize the state space	a large ideal Fermi gas is many-body without interaction-induced correlations
symmetry and integrability	special sectors or conserved quantities can reduce the effective description	$N$ spin-$1/2$ particles restricted to the permutation-symmetric sector span only $N+1$ states
geometry and locality	a chain, fully connected graph, and trapped continuum support different notions of distance and boundary	equal $N$ does not imply equal propagation or finite-size behavior
target observable	a one-point density may converge long before a gap or long-range correlator	one sample can be macroscopic for $n(\mathbf r)$ but not for edge response
desired accuracy and resolution	unresolved levels can look continuous, while high-resolution spectroscopy sees discreteness	“bulk-like” changes when energy or time resolution changes
correlation length	proximity to criticality can make a large sample effectively small	increasing $\xi$ can restore strong finite-size dependence at fixed $L$

A collective description also requires care. If $N$ spins remain in a permutation-symmetric sector, a collective spin $S=N/2$ can be an efficient effective variable. That does not erase the $N$ physical constituents or imply that generic perturbations preserve the sector. It says that symmetry and preparation restrict the dynamically relevant subspace for the question being asked.

Worked Classifications

The point of these examples is not to attach permanent labels to platforms. It is to show how the target question selects the regime.

Three Resonant Bosons in a Trap

• Active degrees: three particle coordinates, reduced to relative and center-of-mass coordinates.
• Target: bound-state ratios and three-body correlations.
• Classification: strongly correlated few-body problem.
• Reason: each constituent and scattering channel remains explicit; universality does not require a thermodynamic limit.

A Four-Site Hubbard Plaquette

• Active degrees: fermionic occupations on four sites.
• Target A: the complete finite spectrum of the isolated plaquette.
• Target B: how a plaquette observable evolves across a family of larger lattices.
• Classification: deliberately context-dependent. Target A can be a small-cluster or few-site problem; Target B treats the plaquette as one finite member of a many-body sequence.

The Hamiltonian alone does not settle the terminology. The family of systems and the claim do.

A 24-Site Interacting Spin Chain

• Active degrees: 24 local spin factors.
• Target: correlations, entanglement, gaps, and response.
• Classification: finite many-body system, not automatically macroscopic and not an infinite chain.
• Evidence boundary: a susceptibility maximum at $L=24$ is a finite-size feature. It does not prove a bulk phase transition.

A Trapped Gas of 100,000 Ideal Fermions

• Active degrees: many occupied single-particle modes constrained by Fermi statistics.
• Target: central density profile at spatial resolution much larger than the interparticle spacing.
• Classification: finite many-body and plausibly macroscopic for that density observable, despite the absence of interactions.
• Caution: the trap is inhomogeneous, so a homogeneous fixed-density thermodynamic limit is a separate construction.

One Impurity in a Fermi Sea

• Active degrees: the impurity plus particle-hole excitations of the bath.
• Target: impurity spectral function or dressing cloud.
• Classification: many-body problem.
• Reason: “one impurity” does not mean one active body when the sea responds dynamically.

Many Spins Confined to a Collective Sector

• Active degrees: $N$ physical spins, with dynamics restricted to a symmetric collective-spin sector.
• Target: collective magnetization under symmetry-preserving evolution.
• Classification: physically many-body with an effective collective description.
• Caution: symmetry-breaking perturbations or local observables can expose degrees of freedom omitted by the reduction.

From Finite Evidence to a Bulk Claim

Finite calculations are indispensable evidence about bulk physics, but the strength of the conclusion must match the evidence.

1. State the exact finite result. Give $L$, geometry, boundary conditions, Hamiltonian parameters, state preparation, observable, and numerical or experimental uncertainty.
2. Define one family of systems. Vary size without silently changing filling, aspect ratio, coupling normalization, temperature, or measurement protocol.
3. Compare several controlled sizes. One large size cannot reveal a trend or distinguish a boundary effect from a bulk effect.
4. Normalize the observable appropriately. Separate total quantities from densities and bulk contributions from boundary contributions.
5. Test competing explanations. Check boundary conditions, crossover scales, truncation error, symmetry sectors, and the possibility of an avoided crossing or pseudocritical peak.
6. Use an explicit extrapolation, bound, or convergence argument. Report the assumed correction form and its stability when sizes or fitting windows change.
7. Limit the conclusion. “Consistent with the expected bulk scaling” is weaker—and often more accurate—than “proves a phase transition.”

For example, replace

The susceptibility peak at $L=24$ proves a phase transition at $g_c$.

with

For open chains of lengths $L=12,16,20,24$, the susceptibility has a size-dependent maximum near $g_c$. Its drift and height are consistent with the proposed critical scaling over this size range; larger sizes, boundary-condition checks, uncertainty estimates, and a controlled extrapolation are still required for a bulk claim.

The physical diagnosis of such size dependence belongs to Finite-Size Effects, the implementation of extrapolations to Finite-Size Scaling in Numerics, and the mathematical target to Thermodynamic Limit.

Canonical Handoffs

Use this page to classify the problem and the strength of the claim. Then continue to the page that owns the technical question.

Question	Canonical destination
How fast does the physical state space grow?	Scaling of Hilbert Space
Which support, range, and decay assumptions make the model local?	Locality in Many-Body Systems
What sequence defines an infinite system?	Thermodynamic Limit
Which physical mechanism makes the accessible system size dependent?	Finite-Size Effects
How are finite-size data extrapolated numerically?	Finite-Size Scaling in Numerics
How do exchange symmetry and indistinguishability change the description?	Identical Particles and Exchange Symmetry
How are variable-particle-number sectors organized?	Fock Space and Occupation Number

The chapter’s scaling sequence now culminates in Emergence and Effective Degrees of Freedom, which audits how a finite or bulk microscopic problem can acquire a predictive effective variable set.

Common Pitfalls

• Many-body means interacting. An ideal quantum gas is already many-body because particle statistics and occupations involve many active constituents.
• Few-body means easy. Resonant three- and four-body problems can contain deep analytical and numerical structure.
• A two-body Hamiltonian describes a two-body system. It may instead be a sum of pair interactions acting on a macroscopic number of particles.
• A few-body observable makes the state few-body. Local correlators are often measured precisely because the full state is many-body.
• Macroscopic means infinite. Every laboratory sample is finite, and different observables approach bulk behavior at different rates.
• A large Hilbert space proves generic computational hardness. Symmetry, integrability, tensor structure, low entanglement, or a restricted sector can make special questions tractable.
• A smooth finite-size crossover is a phase transition. A bulk transition claim requires a specified limit and controlled evidence.
• Large-$N$ always means many particles. $N$ may count fields, flavors, colors, replicas, or an approximation parameter rather than spatial constituents.

Exercises

Exercise 1: Classify the Claim, Not Just the Sample

For each case, state the problem regime, physical scale, and claim type.

1. The exact binding energies of three resonant atoms in a harmonic trap.
2. A correlation function in the ground state of a 20-site spin chain.
3. The measured central density of a trapped cloud containing $10^5$ atoms.
4. The statement that the free-energy density converges as $N,V\to\infty$ at fixed $N/V$.

Solution

1. This is a few-body problem, microscopic in physical scale, and an exact or approximate claim about one finite system.
2. This is a finite many-body problem. Whether it is macroscopic for that correlator depends on its support, the correlation length, boundaries, and the requested accuracy. The stated result is finite-system evidence only.
3. The cloud is finite many-body and may be macroscopic for a coarse central-density observable. The measurement is still a finite-system result, not an infinite-system theorem.
4. This is a thermodynamic-limit claim about a specified sequence. It does not describe the size of any single sample.

Exercise 2: Repair an Overclaim

A calculation finds that the heat-capacity maximum sharpens between $L=8$ and $L=12$. Rewrite the conclusion so that it states what has been shown and lists the minimum missing evidence for a thermodynamic transition.

Solution

A defensible conclusion is: “For the specified geometry and boundary conditions, the heat capacity develops a sharper finite-size maximum between $L=8$ and $L=12$.” A bulk claim additionally needs several controlled sizes, consistent normalization and couplings, uncertainty or truncation estimates, the drift and scaling of the maximum, boundary and aspect-ratio checks, and an extrapolation or theoretical convergence argument. A crossover must remain a competing explanation until those checks distinguish it.

Exercise 3: Three Meanings of Two-Body

Distinguish the following statements:

1. $H=\sum_{i<j}V_{ij}$ is a two-body Hamiltonian.
2. $C_{ij}=\langle S_i^zS_j^z\rangle$ is a two-body observable.
3. $H$ acts on exactly two particles.

Solution

Statement 1 says that each interaction term has support on at most two constituents; the full system may contain arbitrarily many constituents. Statement 2 says that the measured operator has support on two sites, even though its expectation value is evaluated in a many-body state. Statement 3 alone specifies a two-particle system. None of the first two statements implies the third.

Exercise 4: One Sample, Two Scale Judgments

An open chain has length $L=100$ and a correlation length $\xi=3$. Compare a one-site observable at the center with an observable localized on the first site. Can either be called bulk-like from $L/\xi$ alone?

Solution

The center is about 50 sites from either boundary, so $d_{\partial}/\xi\gg1$ supports—but does not prove—the expectation that a local central observable is insensitive to the edges. One should still compare sizes and boundary conditions and check for long-range interactions, conserved global constraints, or other slow scales. The first-site observable is an edge observable by construction; a large $L/\xi$ does not turn it into a bulk observable. Thus macroscopicity is relative to support and target, not only to the global ratio $L/\xi$.

References

• P. W. Anderson, “More Is Different,” Science 177, 393–396 (1972), doi:10.1126/science.177.4047.393.
• E. Braaten and H.-W. Hammer, “Universality in Few-Body Systems with Large Scattering Length,” Physics Reports 428, 259–390 (2006), doi:10.1016/j.physrep.2006.03.001.
• M. Campostrini, A. Pelissetto, and E. Vicari, “Finite-Size Scaling at Quantum Transitions,” Physical Review B 89, 094516 (2014), doi:10.1103/PhysRevB.89.094516.
• P. Coleman, Introduction to Many-Body Physics, Cambridge University Press (2015), doi:10.1017/CBO9781139020916.
• R. K. Pathria and P. D. Beale, Statistical Mechanics, 3rd ed., Elsevier (2011), doi:10.1016/C2009-0-62310-2.