Pith. Sign in

REVIEW

Two and three-point functions in Liouville theory

High Energy Physics - Theory · Submitted 23 Mar 1994 · 19 references

Pith's verdict: The pre-DOZZ construction of generic-β Liouville two- and three-point functions: the closed forms are right in hindsight, but the continuation's normalization and α↔2/α duality are never independently pinned down.

T0 · AI review

How this review was made: An AI referee (selfhosted/deepseek-v4.1-flash) read the paper on October 29, 2026 and answered a fixed list of questions about it. T0 is the starting level every paper gets. The levels describe how deeply Pith has checked the paper, never the authors.

  1. T0 AI review An AI read the paper and answered a fixed list of questions.
  2. T1 Audit Automatic checks that the references and files the paper relies on exist and match.
  3. T2 Compute The paper's code or calculations re-run with the same results.
  4. T3 Formal statement The claims written out in a formal math language a computer can read.
  5. T4 Proof Those claims proved, and the proof checked by computer.

How Pith checks papers →

· 3 major objections · 6 minor · Ask about this paper · The compiled record · Read the full review →
Challenge this review Comment (0) Read on arXiv ↗
Cite ▾
arXiv IDhep-th/9403141 (version 3)
Pith IDTMFKIWJS, Pith's permanent record of this paper and its review
CitationH. Dorn, H.-J. Otto. "Two and three-point functions in Liouville theory." arXiv:hep-th/9403141 (1994). https://pith.science/paper/TMFKIWJS
Cite this reviewBibTeX for Pith's review of this paper
Claim paper

Why Pith says this

Highlight any text to comment on it.

Referee report
The paper

The paper claims

The paper constructs closed-form two- and three-point functions for Liouville exponentials with arbitrary real exponent, and checks the three-point function against the Liouville equation of motion.

The reading

Liouville theory supplies the Weyl degree of freedom of a two-dimensional metric, so its correlation functions are what one needs to dress vertex operators in noncritical string theory and in conformal models beyond the minimal series. The obstacle is that the exponential $e^{βφ}$ is only a free-field construct for special rational values of $β$; for generic real $β$ the operator is an infinite series and no direct computation was available. The authors take the Goulian–Li continuation — treat the number $s$ of cosmological-term insertions as a variable and continue it away from integers — strip it of the degenerate-kinematics restrictions used in the original derivation, and obtain explicit expressions for the three-point amplitude $A_3(β_1,β_2,β_3)$ and the two-point amplitude $A_2(β)$. The evidence offered is a functional identity: the three-point function of $e^{β_2φ}e^{β_3φ}$ with one insertion of $∂^2φ$ equals $αμ^2/2$ times the three-point function with $e^{αφ}$ in that slot, which is the Liouville equation of motion at the level of correlators. If the construction is right, one has for the first time a usable closed form for the gravitational dressing of two- and three-point functions at arbitrary real dressing charge, plus a concrete proposal for where the mass shell of a noncritically dressed tachyon sits.

What carries the argument

The engine is the continued function $f(a,b|s)$, defined for integer $s$ as $Σ_{j=0}^{s-1}Γ(a+bj)$ and extended to arbitrary complex arguments by an integral representation. It obeys four functional relations — shift in $s$, shift in $a$, shift by $b$, and a reflection $b→-b$ — which let the authors move gamma-function arguments around without ever leaving the continued domain. The poles and zeros of $e^{f}$ are then read off from those relations: poles at $a = -bj-ℓ$ and zeros at $a+bs = -bj-ℓ$ for nonnegative integers $j,ℓ$. This is what makes the otherwise formal continuation controllable, and it is what produces the pole-zero spectrum of $A_2$ and $A_3$ that the paper later uses to argue about mass shells. A second piece of machinery is the treatment of the residual Möbius volume: the two-punctured sphere has one conformal Killing vector, so the naive $β_3→0$ limit of $A_3$ diverges, and the authors divide by the explicit volume $V^{(2)}_{CKV}$ of the subgroup fixing $z_1,z_2$ to obtain a finite $A_2$.

Extended reading notes

Core claim

The paper's central claim is that the two- and three-point functions of Liouville exponentials $e^{β_jφ}$ exist in closed form for generic real $β_j$. The three-point function is reduced, by a Möbius transformation argument, to the standard conformal structure with a $β$-dependent amplitude $A_3$ given by a Dotsenko–Fateev integral over $s_3 = (Q-Σβ_j)/α$ screening insertions; evaluating that integral and continuing in $s_3$ produces $A_3$ as a product of gamma functions times three factors $F_i$ built from a continued sum $f(a,b|s)$ of log-gamma terms. The two-point function follows either by taking a $β_3→0$ limit of $A_3$ with the volume of the residual Möbius subgroup divided out, or by differentiating the two-point function with respect to $μ^2$ and matching against the once-integrated three-point function; both routes give the compact expression for $A_2(β)$ in eq. (33). The load-bearing check is eq. (42): the authors verify that $4(Δ_2-Δ_3)^2 ∂_{β_1}A_3|_{β_1=0}$ equals $αμ^2/2$ times $A_3(α,β_2,β_3)$, which is precisely the Liouville equation of motion $∂^2φ = (αμ^2/2)e^{αφ}$ promoted to a statement about three-point functions. They read this as strong evidence that the continuation procedure is the right one, since a uniqueness proof was previously available only for the rational coefficients needed to dress minimal models.

Load-bearing premise

The whole result rests on continuing a sum of log-gamma terms from integer to arbitrary complex values, and on one convention for the otherwise meaningless product $Γ(0)sin(πs)$ — a sign choice the authors say is fixed by requiring the Liouville-equation check to come out right, rather than derived independently.

Editorial extensions

If this is right

  • Gravitational dressing factors for two- and three-point functions become available at arbitrary real dressing charge, not only at the rational values tied to minimal models.
  • The pole-zero structure of $A_2(β)$ yields a candidate mass-shell condition: the tachyon zero at $β = Q/2$ gives $(k-P/2)^2 = (D-1)/12$, with the graviton case shifted to $(k-P/2)^2 = (D-25)/12$.
  • All zeros of $A_2$ other than $β=Q/2$ lie above $Q/2$ and correspond to wave functions that are not normalizable as $φ→+∞$, so they are excluded and the microscopic puncture sits on the boundary of the allowed set.
  • Because $A_3$ vanishes whenever any one of the three $β_j$ is taken from the zero table of $A_2$, and no configuration of three coinciding poles can rescue it, the three-point string $S$-matrix element vanishes for on-shell external states — the property factorization of higher $S$-matrix elements would need.
  • The quasiclassical limit $Q→∞$ reproduces the previously known structure of the normalized two-point function, $G_2/Z ∝ (2b-1)^{-1}(-m^2/8)^{-2b}|z_1-z_2|^{-4b}$, now in explicit form.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The Liouville-equation check is a one-parameter test of a two-parameter continuation: it fixes the ambiguous $\sin(πs)Γ(0)$ convention but cannot by itself confirm that the $β$-dependence away from $β_1=0$ is right, so the natural next test is a four-point function or an independent canonical free-field computation of $A_2$.
  • The mass-shell conjecture is stated in $D$ target-space dimensions rather than $D+1$, and the authors themselves note that the $(D+1)$-dimensional reading of the on-shell condition works cleanly only at $μ=0$; whether the conjecture survives at $μ≠0$ is decidable once the special four-point cases are computed.
  • If the mass-shell zeros of $A_2$ are the right on-shell condition, then the tower of excluded zeros above $β=Q/2$ is a prediction about which would-be states are absent from the noncritical spectrum — a statement that a matrix-model or continuum computation at $c_M<1$ could in principle check.
  • The failure of the one-point function to be well defined in this scheme, which the authors trace to the absence of an $SL(2,\mathbb{C})$-invariant vacuum, suggests that any correlator of the Liouville field itself must be defined with at least two spectator exponentials — a structural constraint that would carry over to any alternative construction of $φ$ correlators.
Original abstract

Based on our generalization of the Goulian-Li continuation in the power of the 2D cosmological term we construct the two and three-point correlation functions for Liouville exponentials with generic real coefficients. As a strong argument in favour of the procedure we prove the Liouville equation of motion on the level of three-point functions. The analytical structure of the correlation functions as well as some of its consequences for string theory are discussed. This includes a conjecture on the mass shell condition for excitations of noncritical strings. We also make a comment concerning the correlation functions of the Liouville field itself.

The compiled record

The paper's own sentences, arranged, kept beside the review and separate from it. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.

Compiled by classifier-1.13-20260917 via OpenRouter, question bank abstract-p1-20260923, 2026-09-23: 1 request, 4,198 input tokens, $0.0002 reported.

What the abstract itself states, as probabilities: claims first 0.05 · states limitation 0.15 · human participants 0.01 · controlled comparison 0.01 · randomized assignment 0.01 · states quantitative result 0.06 · compares with prior results 0.24 · mentions code or data availability 0.01.

The abstract, sentence by sentence

arXiv listing abstract v3. Results and the main claim are underlined; .

Based on our generalization of the Goulian-Li continuation in the power of the 2D cosmological term we construct the two and three-point correlation functions for Liouville exponentials with generic real coefficients. As a strong argument in favour of the procedure we prove the Liouville equation of motion on the level of three-point functions. The analytical structure of the correlation functions as well as some of its consequences for string theory are discussed. This includes a conjecture on the mass shell condition for excitations of noncritical strings. We also make a comment concerning the correlation functions of the Liouville field itself.

background contribution method result interpretation main claim

What the classifier reports for each sentence. The number is the probability it gave the label.
#sentencerolethe authors' own workstates the main conclusionstrength
s0 Based on our generalization of the Goulian-Li continuation in the power of the 2D cosmological term we construct the two and three-point correlation functions for Liouville exponentials with generic real coefficients. contribution 0.90 0.95 0.49 asserted 1.00
s1 As a strong argument in favour of the procedure we prove the Liouville equation of motion on the level of three-point functions. result 0.81 0.95 0.25 asserted 0.97
s2 The analytical structure of the correlation functions as well as some of its consequences for string theory are discussed. contribution 0.74 0.81 0.10 asserted 0.81
s3 This includes a conjecture on the mass shell condition for excitations of noncritical strings. interpretation 0.51 0.82 0.10 speculative 0.96
s4 We also make a comment concerning the correlation functions of the Liouville field itself. contribution 0.90 0.83 0.04 unclear 0.76

What kind of paper the classifier reports this is: theoretical result 0.99 · evidence from proof or derivation 1.00.

What was examined, and what was not

sources read
  • arXiv listing abstract v3
not examined
  • full text
  • figures and tables
  • references and citing sentences
  • relations to other papers
levels run
  • P abstract
not run
  • S sections
  • T body sentences
  • C citations
  • R cross-paper relations

Every quoted sentence on this record is the authors'. Every label is from Pith's closed vocabulary, with the probability shown beside it. Every number in the measurements table is the authors', checked by code. Record as JSON

References
How later papers used it

179 later papers cite it

Evidence: substantial. 178 of the citing papers share no author with it, and 22 of those carry a confident label from Pith's citation classifier.

Pith read 11 of the 255 citing sentences, choosing only those whose citation markers point at this paper alone, and wrote the paraphrases from them. The summary, the counts, the evidence level, the links and the quotes come from code; the summary counts the papers listed below. A later paper is listed under a label when Pith's citation classifier gives one of its sentences that label at probability 0.8 or more (0.95 for support), Pith's reading finds that the citing paper itself did it, and a second check confirms the paraphrase against the sentence. This section lists uses only. It does not list criticism or disagreement, so their absence here says nothing about how the paper has been received. Papers that share an author with this one are listed apart, under "By the same authors". A quote is the citing authors' own sentence, shown when their paper's licence is CC BY, CC BY-SA or CC0. JSON

Uses its method or tool 2 of 9 papers with this label
Compares with it 1 of 11 papers with this label
  • Found that the integral giving the amplitude between its states matches exactly with the b to 0 limit of this paper's DOZZ formula.

    Solving the Schwarzian via the Conformal Bootstrap (2017) · 0.93 · paraphrased: the licence does not allow quoting

How the literature uses it

161 arXiv papers cite this in 228 sentences

Each citing sentence is taken from the citing paper's LaTeX source and typed by Pith's classifier. A label appears at probability 0.8 or more (0.95 for support), a contest at 0.7 or more; the quoted words are the citing authors'. How each label was checked · JSON

Papers citing it · 152 sentences without a confident label

Uses its method or tool 9 papers
  • In order to do this we are going to use the DOZZ formulation of Liouville theory [ref] .

    Bootstrapping closed string field theory · Classical bootstrap for string vertices · 0.99

  • Nevertheless, we can leverage the exact solution of the worldsheet CFT ([ref]) [ref] to deduce the analytic structure of the string amplitudes viewed as analytic functions of the external momenta.

    The complex Liouville string: the worldsheet · Introduction · 0.92

  • At each point in the base, we then perform the integration along the fiber: [equation] where in going from the second to the third line we used the Verlinde inner product [ref] between Liouville blocks [equation] and used the compact notation [ref] [equation] $𝒞$ being a choice of channel decomposition, $𝐏$ being the internal momenta, $ρ_0$ being the Cardy density, and $C_0$ being the DOZZ formula [ref] up to some normalization [ref] .

    On random matrix statistics of 3d gravity · Other amplitudes · 0.93

Uses its data or values 1 paper
Compares with it 10 papers
  • The final result reads [equation] where $Δ_i = η_i (1-η_i)/b^2$ and $C_123$ is given by [equation] with [equation] This matches with the classical limit of the DOZZ formula [ref] .

    Liouville Theory, AdS$_2$ String, and Three-Point Functions · Solving the functional equation · 0.99

  • By comparing to the exact correlation functions, known from [ref] , the authors of [ref] identified the saddles that contribute to the corresponding functional integrals.

    Semi-Classical Analysis of the String Theory Cigar · Introduction and Overview · 0.88

  • Incorporating these leg factors, we finally get f(_1)f(_2)/f(^∗_3) F__1,_2,_3 F__1^∗,_2^∗,_3^∗ ∼'_0 (2_1)(2_2)(2_3)/( _1+_2+_3 -Q)(_2+_3-_1)(_3+_1-_2)(_1+_2-_3) , where [equation] This is precisely the DOZZ formula [ref] apart from the extra normalization factor [equation] Notice that this normalization factor depends on the coefficient $$ of the exponential term $∫d^2z e^2bφ$ in the Liouville action which can be shifted away by shifting $→+$ const, and is therefore clearly scheme-depedent.

    Non-Perturbative Topological Strings And Conformal Blocks · Three-Point Functions · 0.91

Cites it as support 1 paper
  • However, the OPE in the unphysical range of the Liouville momentum formally makes sense, and indeed it was even used in the original derivation of the three-point function [ref] [ref] [ref] .

    Refined Cigar and Omega-deformed Conifold · Refined cigar and $Ω$-deformation · 0.96

Cites it as background 48 papers
  • These partition functions were until recently [ref] unknown, with the sole exception of the W$_2$ case, i.e. the Liouville case, whose three point structure constants are given by the famous DOZZ formula [ref] .

    Toda 3-Point Functions From Topological Strings · Introduction · 0.81

  • The spacelike structure constants $C(a_1 , a_2, a_3)$ are given by the Dorn–Otto–Zamolodchikov–Zamolodchikov (DOZZ) formula [ref] (see [ref] for more discussions).

    Quantum Gravity from Timelike Liouville theory · Conformal bootstrap for Liouville · 0.83

  • The Virasoro primaries of the $c=25$ Liouville CFT are scalar operators $V_P$ labeled by their Liouville momenta $P≥0$, of scaling dimensions $Δ_P=2+2P^2$, subject to the normalization ⟨V_P_1(z,z̅)V_P_2(0)⟩=πδ(P_1-P_2)/|z|^2Δ_P_1 Their structure constants are given by the DOZZ formula [ref] .[Explicitly, in our normalization convention, the $c=25$ Liouville structure constants are given in (2.9) of [ref] .] The closed string asymptotic states, as insertions on the worldsheet, are given by the BRST cohomology representatives of the form V^±_ω= g_s cc̃ e^±i ωX^0 V_P=ω/2, where the superscript $+$ and $-$ corresponds to in- and out-states, respectively.

    Multi-Instanton Calculus in c = 1 String Theory · Multiple ZZ-instantons from the worldsheet perspective · 0.81

Discussion

Disagree with the review, or know something it missed? Leave a comment, highlight any passage above to comment on it, or challenge this review formally.