Open String Veneziano Amplitude

• The open string Veneziano amplitude is a tree-level four-point amplitude expressed via the Euler Beta function with a linear Regge trajectory.
• It exhibits a detailed pole expansion that encodes tachyon, massless gauge boson, and massive Regge excitation contributions in its kinematics.
• It provides an exact benchmark for matching effective field theory interactions, guiding the extraction of cubic and quartic contact terms from string theory.

Searching arXiv for recent and foundational papers on the open-string Veneziano amplitude and closely related developments. The open string Veneziano amplitude is the tree-level four-point open-string amplitude written in Euler Beta-function form. In the open bosonic tachyon sector it appears as

$$A(s,t)=B(-\alpha(s),-\alpha(t)) =\frac{\Gamma(-\alpha(s))\Gamma(-\alpha(t))}{\Gamma(-\alpha(s)-\alpha(t))},$$

with linear Regge trajectory $\alpha(s)=1+\frac{s}{2}$ in one standard normalization, so that $s+t+u=-8$ for four external tachyons of mass squared $-2$ (He et al., 2015). In another common normalization,

$${\cal A}_{Ven}(s,t)=g_o^2\frac{\Gamma(-\ell_s^2 s-1)\Gamma(-\ell_s^2 t-1)}{\Gamma(-\ell_s^2 s-\ell_s^2 t-2)},$$

which is the bosonic open-string tachyon baseline amplitude (Firrotta, 2024). For the maximally supersymmetric color-ordered open superstring, the corresponding four-point amplitude is

$$A^{\rm str}(s,u)=-(\alpha' s)^2\, \frac{\Gamma(-\alpha' s)\Gamma(-\alpha' u)}{\Gamma\!\left(1-\alpha'(s+u)\right)},$$

so the beta-function kernel survives, but with the supersymmetric kinematic prefactor $s^2$ (Berman et al., 2024).

1. Canonical forms and kinematics

In the bosonic open-string tachyon amplitude, the Mandelstam variables can be written as

$$s=-(k_1+k_2)^2,\qquad t=-(k_2+k_3)^2,\qquad u=-(k_1+k_3)^2,$$

with

$$\alpha(s)=1+\frac{s}{2},\qquad \alpha(s)+\alpha(t)+\alpha(u)=-1,$$

in the normalization where the tachyon mass squared is $-2$ (He et al., 2015). The same amplitude is also written, in the notation of open bosonic string scattering on a $\alpha(s)=1+\frac{s}{2}$0-brane, as

$\alpha(s)=1+\frac{s}{2}$1

with $\alpha(s)=1+\frac{s}{2}$2 and $\alpha(s)=1+\frac{s}{2}$3 (Lee, 2019).

A distinction that recurs throughout the literature is between the single ordered beta-function amplitude and crossing-symmetrized combinations. For the bosonic tachyon amplitude,

$\alpha(s)=1+\frac{s}{2}$4

and the symmetrized object plays a special role in reformulations involving zeta functions (He et al., 2015). In the superstring case, the color-ordered amplitude is instead organized as a single partial amplitude with Chan–Paton traces, and the $\alpha(s)=1+\frac{s}{2}$5 prefactor removes the would-be $\alpha(s)=1+\frac{s}{2}$6 pole in the scalar component that is bootstrapped in ten dimensions (Berman et al., 2024).

The disk worldsheet representation is the common starting point. With open-string insertions on the real line at $\alpha(s)=1+\frac{s}{2}$7, the flat-space integrand is

$\alpha(s)=1+\frac{s}{2}$8

which evaluates to the Veneziano beta function in the conventions used for AdS deformations (Alday et al., 2024).

2. Pole expansion, duality, and factorization

The defining structural feature of the open string Veneziano amplitude is its pole expansion. For the four-tachyon bosonic amplitude,

$\alpha(s)=1+\frac{s}{2}$9

and the first terms are

$s+t+u=-8$0

The first pole at $s+t+u=-8$1 is the tachyon exchange, the pole at $s+t+u=-8$2 is the massless gauge-boson exchange, and the higher poles at $s+t+u=-8$3 encode the infinite tower of massive Regge excitations (Lee, 2019).

The same factorization pattern persists for arbitrary external open-string excitations. In the coherent-state formalism for four arbitrary bosonic open-string states,

$s+t+u=-8$4

so the standard Beta-function kernel is unchanged while the external-state dependence is carried by a universal dressing operator (Firrotta, 2024). The ordinary Veneziano residue structure then reappears in the factorization formula

$s+t+u=-8$5

which is the direct arbitrary-state extension of Veneziano factorization (Firrotta, 2024).

Partial-wave positivity gives a direct amplitude-level formulation of unitarity. For

$s+t+u=-8$6

the residue at $s+t+u=-8$7 is expanded as

$s+t+u=-8$8

and positivity means $s+t+u=-8$9 for all $-2$0 (Mansfield, 27 Feb 2025). For the open superstring case this is proved directly for all $-2$1; the explicit example

$-2$2

shows why the bound is sharp (Mansfield, 27 Feb 2025).

3. Worldsheet realization and alternative derivations

At tree level, the amplitude is the ordered open-string disk integral, and many later developments preserve that form even when the derivation changes. One direct on-shell reconstruction uses BCFW recursion. With a $-2$3 shift, the four-tachyon bosonic amplitude is written as a sum over the full tower of intermediate open-string states,

$-2$4

and the fixed-level residue resums to

$-2$5

The resulting pole series

$-2$6

is identified with $-2$7 (Fotopoulos, 2010).

A different derivation fixes one $-2$8 degree of freedom by inserting the mostly BRST exact operator

$-2$9

so that a five-point correlator of four tachyons plus $${\cal A}_{Ven}(s,t)=g_o^2\frac{\Gamma(-\ell_s^2 s-1)\Gamma(-\ell_s^2 t-1)}{\Gamma(-\ell_s^2 s-\ell_s^2 t-2)},$$0 reproduces the Veneziano amplitude up to the sign factor

$${\cal A}_{Ven}(s,t)=g_o^2\frac{\Gamma(-\ell_s^2 s-1)\Gamma(-\ell_s^2 t-1)}{\Gamma(-\ell_s^2 s-\ell_s^2 t-2)},$$1

which is interpreted as a signed intersection number (Kishimoto et al., 2021).

The “novel string field theory” of objects sought to derive the same amplitude from overlaps of initial and final object states. In the 2014 construction the explicit derivation reached only one channel contribution, $${\cal A}_{Ven}(s,t)=g_o^2\frac{\Gamma(-\ell_s^2 s-1)\Gamma(-\ell_s^2 t-1)}{\Gamma(-\ell_s^2 s-\ell_s^2 t-2)},$$2, and the full crossing-symmetric amplitude was deferred (Nielsen et al., 2014). The later extension allowed negative-energy constituents with effectively

$${\cal A}_{Ven}(s,t)=g_o^2\frac{\Gamma(-\ell_s^2 s-1)\Gamma(-\ell_s^2 t-1)}{\Gamma(-\ell_s^2 s-\ell_s^2 t-2)},$$3

so that compensating chain segments can annihilate, and this was proposed as the mechanism restoring all three Veneziano terms (Nielsen et al., 2017).

4. Effective action and tachyon-sector matching

The open string Veneziano amplitude can also be used as an exact benchmark for low-energy effective field theory. In the open bosonic string on a single $${\cal A}_{Ven}(s,t)=g_o^2\frac{\Gamma(-\ell_s^2 s-1)\Gamma(-\ell_s^2 t-1)}{\Gamma(-\ell_s^2 s-\ell_s^2 t-2)},$$4-brane, the cubic tachyon coupling extracted from the three-string amplitude is

$${\cal A}_{Ven}(s,t)=g_o^2\frac{\Gamma(-\ell_s^2 s-1)\Gamma(-\ell_s^2 t-1)}{\Gamma(-\ell_s^2 s-\ell_s^2 t-2)},$$5

and this reproduces exactly the first Veneziano pole, the tachyon exchange term $${\cal A}_{Ven}(s,t)=g_o^2\frac{\Gamma(-\ell_s^2 s-1)\Gamma(-\ell_s^2 t-1)}{\Gamma(-\ell_s^2 s-\ell_s^2 t-2)},$$6 (Lee, 2019).

The next pole, however,

$${\cal A}_{Ven}(s,t)=g_o^2\frac{\Gamma(-\ell_s^2 s-1)\Gamma(-\ell_s^2 t-1)}{\Gamma(-\ell_s^2 s-\ell_s^2 t-2)},$$7

requires gauge-boson exchange. In the single-brane Abelian setup, a real tachyon field gives a vanishing $${\cal A}_{Ven}(s,t)=g_o^2\frac{\Gamma(-\ell_s^2 s-1)\Gamma(-\ell_s^2 t-1)}{\Gamma(-\ell_s^2 s-\ell_s^2 t-2)},$$8 coupling, so the exact string amplitude cannot be matched unless the tachyon is represented as a complex field (Lee, 2019). With

$${\cal A}_{Ven}(s,t)=g_o^2\frac{\Gamma(-\ell_s^2 s-1)\Gamma(-\ell_s^2 t-1)}{\Gamma(-\ell_s^2 s-\ell_s^2 t-2)},$$9

the perturbative gauge exchange still fails to reproduce the full Veneziano contribution, and the remaining mismatch is local: $$A^{\rm str}(s,u)=-(\alpha' s)^2\, \frac{\Gamma(-\alpha' s)\Gamma(-\alpha' u)}{\Gamma\!\left(1-\alpha'(s+u)\right)},$$0 This quartic interaction is therefore fixed by matching the exact four-tachyon Veneziano amplitude to the local field theory (Lee, 2019).

The resulting tachyon potential,

$$A^{\rm str}(s,u)=-(\alpha' s)^2\, \frac{\Gamma(-\alpha' s)\Gamma(-\alpha' u)}{\Gamma\!\left(1-\alpha'(s+u)\right)},$$1

has a stable minimum, and the paper argues that tachyon condensation makes both tachyon and gauge fields massive at the string/Planck scale in the author’s conventions (Lee, 2019). In this use of the amplitude, the Veneziano function is not merely a scattering formula; it is the exact four-point datum that fixes contact terms in the effective action.

5. Positivity, bootstrap, and asymptotic control

A recurring theme is that the Veneziano amplitude is both highly constrained and easy to misuse. One common misconception is to treat asymptotic uniqueness as an exact derivation of the full amplitude. In fact, the weakly-coupled higher-spin analysis proves only that for large positive $$A^{\rm str}(s,u)=-(\alpha' s)^2\, \frac{\Gamma(-\alpha' s)\Gamma(-\alpha' u)}{\Gamma\!\left(1-\alpha'(s+u)\right)},$$2,

$$A^{\rm str}(s,u)=-(\alpha' s)^2\, \frac{\Gamma(-\alpha' s)\Gamma(-\alpha' u)}{\Gamma\!\left(1-\alpha'(s+u)\right)},$$3

and explicitly does not claim exact equality at finite kinematics (Caron-Huot et al., 2016).

The first universal correction to this asymptotic in weakly interacting higher-spin theories is also known: $$A^{\rm str}(s,u)=-(\alpha' s)^2\, \frac{\Gamma(-\alpha' s)\Gamma(-\alpha' u)}{\Gamma\!\left(1-\alpha'(s+u)\right)},$$4 where $$A^{\rm str}(s,u)=-(\alpha' s)^2\, \frac{\Gamma(-\alpha' s)\Gamma(-\alpha' u)}{\Gamma\!\left(1-\alpha'(s+u)\right)},$$5 is the complete elliptic integral of the first kind (Sever et al., 2017). This correction is universal in that high-energy imaginary-angle regime, but it is not a correction to the exact tree-level Veneziano amplitude itself (Sever et al., 2017).

Bottom-up bootstrap methods recover a complementary part of the structure. In ten dimensions, the scalar component of the color-ordered open-superstring amplitude

$$A^{\rm str}(s,u)=-(\alpha' s)^2\, \frac{\Gamma(-\alpha' s)\Gamma(-\alpha' u)}{\Gamma\!\left(1-\alpha'(s+u)\right)},$$6

has low-energy coefficients

$$A^{\rm str}(s,u)=-(\alpha' s)^2\, \frac{\Gamma(-\alpha' s)\Gamma(-\alpha' u)}{\Gamma\!\left(1-\alpha'(s+u)\right)},$$7

and the bootstrap shows that if there is only a single state at the lowest mass, it must be a scalar; with the ratio $$A^{\rm str}(s,u)=-(\alpha' s)^2\, \frac{\Gamma(-\alpha' s)\Gamma(-\alpha' u)}{\Gamma\!\left(1-\alpha'(s+u)\right)},$$8 fixed to its string value, the second state is forced to appear at $$A^{\rm str}(s,u)=-(\alpha' s)^2\, \frac{\Gamma(-\alpha' s)\Gamma(-\alpha' u)}{\Gamma\!\left(1-\alpha'(s+u)\right)},$$9 and to be a vector, and the Veneziano amplitude is isolated on a small island in parameter space (Berman et al., 2024).

Positivity, bootstrap, and asymptotic analysis therefore constrain different aspects of the same object. Positivity proves consistent factorization in $s^2$0 (Mansfield, 27 Feb 2025); bootstrap reconstructs low-energy data and the first few massive states (Berman et al., 2024); asymptotic uniqueness fixes the large positive $s^2$1 logarithm but not the finite-$s^2$2 amplitude (Caron-Huot et al., 2016).

6. Generalizations, deformations, and curved-space analogues

Several extensions preserve the Veneziano kernel while modifying its interpretation or dressing.

For higher points, the five-point open-string tachyon amplitude can be reorganized as

$s^2$3

so the four-point Veneziano amplitude becomes the building block for closed-form five-point formulas (Stoica, 2021). For arbitrary excited bosonic open strings, the four-point amplitude keeps the same Beta-function core and acquires a coherent-state dressing built from DDF data and Jacobi-polynomial structures (Firrotta, 2024).

A different direction is $s^2$4-deformation. The Coon-type deformation is

$s^2$5

which reduces to the ordinary Veneziano amplitude as $s^2$6 and flows to scalar field-theory scattering as $s^2$7 (Li et al., 2023). Closely related work emphasizes that the Coon amplitude is a deformation of the standard Veneziano amplitude with an accumulation point in the spectrum, and that for $s^2$8 it agrees with the low-energy Veneziano amplitude (Maldacena et al., 2022).

The flat-space amplitude also admits precise AdS deformations. In the D7-brane setup of type IIB on $s^2$9, the AdS amplitude is written as

$$s=-(k_1+k_2)^2,\qquad t=-(k_2+k_3)^2,\qquad u=-(k_1+k_3)^2,$$0

so the ordinary open-string Veneziano integrand is dressed by curvature-dependent kernels built from single-valued multiple polylogarithms on the real line (Alday et al., 2024). In the small-curvature expansion, the first correction

$$s=-(k_1+k_2)^2,\qquad t=-(k_2+k_3)^2,\qquad u=-(k_1+k_3)^2,$$1

is fixed by combining a dispersion relation in the dual $$s=-(k_1+k_2)^2,\qquad t=-(k_2+k_3)^2,\qquad u=-(k_1+k_3)^2,$$2 SCFT with a worldsheet ansatz in MPLs (Alday et al., 2024). The same AdS setting supports monodromy relations deforming the flat-space ones: $$s=-(k_1+k_2)^2,\qquad t=-(k_2+k_3)^2,\qquad u=-(k_1+k_3)^2,$$3 and these reduce to the usual monodromy relations of the flat-space Veneziano amplitude when curvature is removed (Alday et al., 2 Sep 2025).

Finally, the crossing-symmetrized bosonic four-point amplitude admits an exact zeta-function rewriting,

$$s=-(k_1+k_2)^2,\qquad t=-(k_2+k_3)^2,\qquad u=-(k_1+k_3)^2,$$4

but this identity applies only to the fully symmetrized amplitude; the unsymmetrized Beta-function amplitude cannot be rewritten purely in terms of zeta functions (He et al., 2015). This distinction is representative of a broader point: many modern reformulations preserve the Veneziano structure only after a specific choice of ordering, symmetrization, or kinematic regime has been fixed.