Majda's no-shock conjecture for one-dimensional linearly degenerate quasilinear hyperbolic systems: ODE-type amplitude blowup versus shock formation

For one-dimensional first-order quasilinear hyperbolic systems whose characteristic fields are all linearly degenerate in Lax's sense, Majda's 1984 conjecture asserts that finite-time breakdown of a classical solution, when it occurs, is never of shock type: the amplitude ($C^0$ norm) of the solution must blow up together with its derivatives, i.e. the singularity is of ODE type. The conjecture has been verified only for structurally restricted classes (block-structured, part-rich, completely reducible, diagonal, or specific physical systems such as the Kerr–Debye system). The open problem is to prove or refute the conjecture for general strictly hyperbolic linearly degenerate systems, with no block structure, part richness, complete reducibility, or diagonality imposed and the eigenvector frame allowed to be fully state-dependent, thereby completing the classification of singularity mechanisms in 1-D quasilinear hyperbolic systems: genuinely nonlinear eigenvalue variation produces shocks, while eigenvector-induced coupling must produce only ODE-type amplitude blowup.

Background

A one-dimensional first-order quasilinear hyperbolic system is a system of the form $u_t + A(u)u_x = F(u)$ for an unknown $\mathbb{R}^n$-valued function $u(t,x)$, where $A(u)$ is a smooth $n\times n$ matrix. The system is strictly hyperbolic if $A(u)$ has $n$ distinct real eigenvalues $\lambda_1(u)<\dots<\lambda_n(u)$; each pair $(\lambda_i, r_i)$ of an eigenvalue and its eigenvector field is a characteristic family, and the curves $dx/dt=\lambda_i(u)$ are the characteristics. Lax's classification splits the families into genuinely nonlinear ones, where the directional derivative of the speed along the eigenvector is nonzero ($\nabla\lambda_i\cdot r_i\neq 0$, so wave speed changes with amplitude), and linearly degenerate ones, where $\nabla\lambda_i(u)\cdot r_i(u)\equiv 0$, so the speed is constant along the integral curves of its own eigenvector field. For genuinely nonlinear systems, the classical results of Lax (2x2 systems), John (n×n genuinely nonlinear systems), and Liu show that smooth finite-amplitude data generically develop a shock in finite time: the solution itself stays bounded while its first derivatives blow up, because faster characteristics behind catch slower ones ahead. For linearly degenerate systems this eigenvalue-driven mechanism is absent, and many physically important systems belong to this class: the Born–Infeld model of nonlinear electrodynamics, the Chaplygin gas (a fluid whose pressure decreases with density and becomes negative, $p=-k/\rho$, so that compression accelerates the flow instead of retarding it), equations of extremal timelike surfaces (surfaces in Minkowski spacetime that extremize the Lorentzian area functional), the Kerr–Debye model in nonlinear optics (in which the nonlinear polarization responds with a finite relaxation time), and recent dissipative relativistic fluid models whose first-order formulations have exclusively linearly degenerate modes (Freistuhler). What remains is the eigenvector mechanism: when the eigenvector frame $r_i(u)$ genuinely rotates with the state, it couples the characteristic families and can amplify the solution's own amplitude, mimicking the blowup of a finite-dimensional ODE such as $w'=w^2$; such ODE-type singularities (amplitude unbounded, or state escaping the domain of definition of the system) are known to occur, e.g. for periodic solutions of linearly degenerate systems (Qu) and as mass concentration in the Chaplygin gas (Kong–Wei–Zhang). In his 1984 monograph, Majda conjectured that linear degeneracy excludes shock formation altogether: a classical solution of a totally linearly degenerate system can break down only by leaving the domain of definition of the system (amplitude blowup), never by gradient blowup at bounded state; Brenier later emphasized the same conjecture in his study of the augmented Born–Infeld equations. In the one-dimensional setting the conjecture takes the sharp form recorded by Li, Peng, Yang and Zhou: if the $C^1$ solution of the Cauchy problem for a strictly hyperbolic system with all linearly degenerate characteristics blows up at a finite time $T_*$, then its $C^0$ norm must also blow up at $T_*$; equivalently, for compactly supported smooth initial data, a uniform $C^0$ bound on $[0,T]$ forces uniform $W^{1,p}$ bounds (bounds on the first spatial derivatives in the $L^p$ norm) for every $1<p\le\infty$ (Qu's energy-method reformulation, under which all $W^{1,p}$ norms blow up simultaneously). The conjecture is known to hold only under structural hypotheses that effectively decouple the eigenvector frame: strictly block-hyperbolic, part-rich, successively block-closed systems (Li–Peng–Yang–Zhou), blocked or completely reducible systems without part richness (Qu), diagonal systems (Yang), and the Kerr–Debye system (Qu); global smooth solutions are also known for special classes such as 2x2 systems in the work of Kong and Tsuji. A multi-dimensional variant (global smooth solutions for totally linearly degenerate symmetric systems) was resolved by Ding, Xin and Yin only for the 2D irrotational Chaplygin gas with short-pulse initial data, and they state it remains open for general small data. Beyond these classes, the eigenvector frame of a general linearly degenerate system cannot be straightened: no Riemann invariants (scalar functions of the state that are constant along the characteristics of their own family and would allow the system to be rewritten in decoupled diagonal form) and no characteristic-coordinate decoupling are available, and the conjecture has remained open since 1984.

Problem Statement

Consider the Cauchy problem for a one-dimensional first-order quasilinear hyperbolic system $u_t + A(u)\,u_x = F(u),\qquad u(0,x)=u_0(x),$ where $u$ takes values in $\mathbb{R}^n$, $A(u)$ is a smooth $n\times n$ matrix, and $F(u)$ is a smooth source term. Assume that $A(u)$ has $n$ distinct real eigenvalues $\lambda_1(u)<\dots<\lambda_n(u)$ (strict hyperbolicity); let $r_i(u)$ denote a smooth eigenvector field of $A(u)$ associated with $\lambda_i(u)$, and let $\nabla$ denote the gradient with respect to the state variable $u$. Assume further that every characteristic field is linearly degenerate in Lax's sense, $\nabla\lambda_i(u)\cdot r_i(u)\equiv 0$ for $i=1,\dots,n$, and that no block structure, part richness, complete reducibility, or diagonality is imposed: the eigenvector frame $r_i(u)$ is allowed to be fully state-dependent. Majda's conjecture (1984) asserts that breakdown of the classical $C^1$ solution, if it occurs at a finite time $T_*$, is never of shock type (no gradient blowup at uniformly bounded state): the $C^0$ norm $\|u(t,\cdot)\|_{C^0}$ must itself blow up as $t\uparrow T_*$; equivalently, for compactly supported smooth initial data, a uniform $C^0$ bound on $[0,T]$ implies uniform $W^{1,p}$ bounds (bounds on the first spatial derivatives in the $L^p$ norm) for every $1<p\le\infty$. Prove or refute this conjecture for the general class above, thereby completing the classification of finite-time singularity mechanisms in 1-D quasilinear hyperbolic systems: genuinely nonlinear eigenvalue variation $\nabla\lambda_i\cdot r_i\neq 0$ produces shock formation (bounded state, gradient blowup), while under full linear degeneracy the eigenvector-frame coupling produces only ODE-type amplitude blowup. A refutation consists of a strictly hyperbolic, linearly degenerate system and smooth initial data whose $C^1$ solution develops unbounded first derivatives at a finite time while the state remains uniformly bounded; a proof must cover the full class above without any structural decoupling of the eigenvector frame.

References

1. P. D. Lax, "Hyperbolic systems of conservation laws II", Communications on Pure and Applied Mathematics 10 (1957) 537–566, DOI 10.1002/cpa.3160100406, https://doi.org/10.1002/cpa.3160100406
2. Fritz John, "Formation of singularities in one-dimensional nonlinear wave propagation", Communications on Pure and Applied Mathematics 27 (1974) 377–405, DOI 10.1002/cpa.3160270307, https://doi.org/10.1002/cpa.3160270307
3. Tai-Ping Liu, "Development of singularities in the nonlinear waves for quasi-linear hyperbolic partial differential equations", Journal of Differential Equations 33 (1979) 92–111, DOI 10.1016/0022-0396(79)90082-2, https://doi.org/10.1016/0022-0396(79)90082-2
4. A. Majda, "Compressible Fluid Flow and Systems of Conservation Laws in Several Space Variables", Applied Mathematical Sciences 53, Springer-Verlag, New York, 1984, ISBN 978-0-387-96037-1, DOI 10.1007/978-1-4612-1116-7, https://doi.org/10.1007/978-1-4612-1116-7
5. De-Xing Kong and Mikio Tsuji, "Global solutions for 2x2 hyperbolic systems with linearly degenerate characteristics", Funkcialaj Ekvacioj 42 (1999) 129–155, zbMATH 2112693 (Zbl 1141.35416), https://zbmath.org/2112693
6. Yann Brenier, "Hydrodynamic Structure of the Augmented Born-Infeld Equations", Archive for Rational Mechanics and Analysis 172 (2004) 65–91, DOI 10.1007/s00205-003-0291-4, https://doi.org/10.1007/s00205-003-0291-4
7. Ta-Tsien Li, Yue-Jun Peng, Yong-Fu Yang and Yi Zhou, "Mechanism of the formation of singularities for quasilinear hyperbolic systems with linearly degenerate characteristic fields", Mathematical Methods in the Applied Sciences 31 (2008) 193–227, DOI 10.1002/mma.902, https://doi.org/10.1002/mma.902 (preprint at https://hal.science/hal-00489220)
8. Yong-Fu Yang, "Mechanism of the formation of singularities for diagonal systems with linearly degenerate characteristic fields", Communications on Pure and Applied Analysis 8 (2009) 757–768, DOI 10.3934/cpaa.2009.8.757, https://doi.org/10.3934/cpaa.2009.8.757
9. Peng Qu, "Mechanism of singularity formation for quasilinear hyperbolic systems with linearly degenerate characteristic fields", Journal of Differential Equations 251 (2011) 2066–2081, DOI 10.1016/j.jde.2011.07.005, https://doi.org/10.1016/j.jde.2011.07.005
10. Peng Qu, "ODE singularity for blocked linearly degenerate quasilinear hyperbolic systems", Mathematical Methods in the Applied Sciences 34 (2011) 950–957, DOI 10.1002/mma.1413, https://doi.org/10.1002/mma.1413
11. Peng Qu, "Blow-up of periodic solutions to linearly degenerate quasilinear hyperbolic systems", Nonlinear Analysis: Theory, Methods & Applications 75 (2012) 6520–6523, DOI 10.1016/j.na.2012.07.027, https://doi.org/10.1016/j.na.2012.07.027
12. De-Xing Kong, Changhua Wei and Qiang Zhang, "Formation of singularities in one-dimensional Chaplygin gas", Journal of Hyperbolic Differential Equations 11 (2014) 521–561, DOI 10.1142/S0219891614500155, https://doi.org/10.1142/S0219891614500155 (preprint arXiv:1311.3474, https://arxiv.org/abs/1311.3474)
13. Heinrich Freistuhler, "Linear Degeneracy in a Class of Nonlinear Second-Order Hyperbolic Systems", arXiv:2406.00719 (2024), https://arxiv.org/abs/2406.00719
14. Bingbing Ding, Zhouping Xin and Huicheng Yin, "On global smooth solutions to the 2D isentropic and irrotational Chaplygin gases with short pulse data", Advances in Mathematics 476 (2025) Article 110362, DOI 10.1016/j.aim.2025.110362, https://doi.org/10.1016/j.aim.2025.110362 (preprint arXiv:2204.05017, https://arxiv.org/abs/2204.05017)
15. Qian Cheng, Shuang-Shuang Duan, Chun-Lei He and Shou-Jun Huang, "Formation of Singularities for Hyperbolic Differential Equations With Geometric Backgrounds", Mathematical Methods in the Applied Sciences 48 (2025) 10352–10360, DOI 10.1002/mma.10882, https://doi.org/10.1002/mma.10882
16. Shunkai Mao and Peng Qu, "Non-uniqueness for the nonlinear dynamical Lamé system", Journal of Differential Equations 446 (2025) Article 113603, DOI 10.1016/j.jde.2025.113603, https://doi.org/10.1016/j.jde.2025.113603 (preprint arXiv:2502.07385, https://arxiv.org/abs/2502.07385)