Antilinear Map¶

An additive map between complex vector spaces that transports scalar multiplication through complex conjugation, so A(lambda v) = conjugate(lambda) A(v) rather than lambda A(v).

Version: v2 · 2026-08-30 · History
Domain-specific #: 1293
Origin domain: mathematics
Subdomain: complex linear algebra
Aliases: Conjugate Linear Map, Conjugate Linear Operator, Semilinear Map with Complex Conjugation

Core Idea¶

An Antilinear Map, also called a conjugate-linear map, is a function \(A:V\to W\) between complex vector spaces satisfying

\[ A(v+w)=A(v)+A(w), \qquad A(\lambda v)=\overline{\lambda}\,A(v) \]

for all \(v,w\in V\) and \(\lambda\in\mathbb C\). Equivalently,

\[ A(\alpha v+\beta w)=\overline{\alpha}A(v)+\overline{\beta}A(w). \]

The map preserves vector addition but twists scalar transport by the distinguished field automorphism \(z\mapsto\overline z\). This is not arbitrary nonlinearity. It is a precise semilinear rule: over the underlying real vector spaces, (A) is linear, while multiplication by (i) reverses sign, (A(iv)=-iA(v)).^[1]^[2]

The locked identity is complex vector-space domain + complex vector-space codomain + additive map + conjugate-homogeneity invariant + explicit conjugation convention. In coordinates, every finite-dimensional antilinear map can be written \(A(z)=M\overline z\) for a complex matrix (M), once bases and coordinatewise conjugation are chosen. The matrix records an ordinary linear action after the conjugation step. Coordinate-free language replaces that choice with the conjugate vector space: an antilinear map \(V\to W\) is the same underlying function as a linear map \(overline V\to W\), or equivalently a linear map \(V\to\overline W\), with the appropriate canonical identifications.^[3]^[2]

Antilinearity is load-bearing in complex inner-product theory, duality, real structures, spinors, and quantum symmetries. Depending on whether an inner product is declared linear in its first or second argument, fixing the other slot yields a linear or antilinear functional; the convention must be stated. In quantum mechanics, time-reversal symmetries are represented antiunitarily: they are antilinear and preserve transition probabilities, with their square carrying physical information.^[4]^[5]

The abstraction survives as domain-specific because its invariant recurs across mathematical domains while remaining inseparable from fields with a specified nontrivial automorphism, especially complex conjugation. Its general structural residue—preserving an operation while twisting another through an automorphism—is semilinearity, not a substrate-independent prime.

Structural Signature¶

the complex domain (V) — an additive group equipped with scalar multiplication by (mathbb C);
the complex codomain (W) — another complex vector space in which outputs and conjugated scalars act;
the map (A) — a total or, in operator theory, possibly densely defined function on a stated linear domain;
additivity — (A(v+w)=A(v)+A(w)) and therefore (A(0)=0), (A(-v)=-A(v));
the conjugation automorphism — \(\lambda\mapsto\overline\lambda\), fixing real scalars and sending (i) to (-i);
conjugate homogeneity — \(A(\lambda v)=\overline\lambda A(v)\);
real linearity — restriction of scalars to (mathbb R) gives (A(av+bw)=aA(v)+bA(w)) for real (a,b);
complex-structure reversal — writing (Jv=iv), antilinearity is (AJ=-JA);
the conjugate-space factorization — (A) becomes ordinary complex-linear after conjugating the scalar structure of one endpoint;
composition parity — linear after linear and antilinear after antilinear are linear; one linear and one antilinear factor yield antilinear composition;
optional analytic structure — norms, boundedness, closedness, domains, adjoints, isometry, or antiunitarity are additional properties, not part of the bare definition.

Recognition test. Verify additivity and test the scalar (i). For an additive real-linear map, (A(iv)=-iA(v)) for every (v) establishes conjugate linearity, while (A(iv)=iA(v)) establishes complex linearity. A map satisfying neither can still be real-linear but is not antilinear.

What It Is Not¶

Not a complex-linear map. Complex linearity preserves (lambda); antilinearity replaces it by \(overline\lambda\). They coincide only when the relevant scalars are real or the map is zero.
Not a generic nonlinear function. Antilinear maps form a tightly controlled real-linear class with a matrix representation and predictable composition rules.
Not every real-linear map between complex spaces. A general real-linear map can contain both a complex-linear part and an antilinear part.
Not an antiunitary operator automatically. Antiunitary additionally requires a bijective norm/inner-product-preserving condition. Antilinear maps can be noninjective, nonisometric, unbounded, or zero.
Not a conjugation operator automatically. In operator theory, a conjugation is usually an antilinear isometric involution (C^2=I). Antilinearity alone supplies neither isometry nor involution.
Not the complex conjugate of a matrix as an operation on matrices. Entrywise conjugation is a related map; a map \(z\mapsto M\overline z\) includes both conjugation and a linear matrix action.
Not the adjoint. Taking \(T\mapsto T^*\) is conjugate-linear on an operator space, while each (T^*) itself is normally a linear operator on vectors. Confusing the level of mapping changes the type.
Not an antiholomorphic map in full. Antilinear maps are special affine-free antiholomorphic maps; general antiholomorphic maps need not be additive or homogeneous.
Not conjugate transpose. The dagger operation combines transpose and complex conjugation and acts on matrices or operators, not necessarily as the vector-space map under discussion.

Scope of Application¶

In finite-dimensional complex linear algebra, coordinatewise conjugation \(K(z)=\overline z\) is the canonical example. Every antilinear \(A:\mathbb C^n\to\mathbb C^m\) has the form (A=M K), hence is determined by a complex matrix once a conjugation associated with bases is chosen. Changing bases transforms that matrix differently from an ordinary linear-operator matrix because the input transition matrix is conjugated.

In functional analysis, continuous antilinear functionals form the anti-dual. Many texts choose inner products linear in the first slot and conjugate-linear in the second, while physics often reverses the convention. The Riesz representation theorem therefore produces a linear or antilinear identification with the dual depending on which dual and slot convention are used. Statements about “the” Riesz map must expose that choice.^[1]^[3]

In operator theory, antilinear operators can be bounded or unbounded, normal or self-adjoint under suitable antilinear-adjoint definitions, and factored using conjugations. Recent systematic treatments use conjugate Hilbert spaces to transport results from linear operators without erasing the altered scalar law.^[2]

In quantum mechanics, Wigner's theorem permits symmetries preserving transition probabilities to be implemented by unitary or antiunitary transformations. Time reversal is the standard antiunitary case. Its antilinearity conjugates amplitudes and reverses (i), which is essential when reversing the sign of time in the Schrödinger equation.^[4]^[5]

In complex geometry and representation theory, antilinear involutions encode real structures: fixed points can recover a real form whose complexification returns the original object. Spinor calculus and complex representations likewise use dotted or conjugate indices to track conjugated representations rather than silently treating them as the same complex-linear space.

Clarity¶

The word anti means “twisted by conjugation,” not “negative.” The map \(z\mapsto-\overline z\) is antilinear, but so is \(z\mapsto\overline z\); sign is irrelevant to the class.

The field matters. Over (mathbb R), conjugation is the identity, so the antilinear law collapses to ordinary linearity. Over a general field (F) with automorphism (sigma), a (sigma)-semilinear map satisfies \(A(\lambda v)=\sigma(\lambda)A(v)\). “Antilinear” conventionally selects \(F=\mathbb C\) and (sigma) equal to complex conjugation.

The conjugate vector space (overline V) has the same underlying additive group as (V) but scalar multiplication \(\lambda\cdot_{\overline V}v=\overline\lambda\,v\). This is why antilinearity is ordinary linearity after a scalar-structure change. It is not a claim that vectors possess basis-independent componentwise bars; componentwise conjugation requires a chosen real structure or basis.

Inner-product convention is another major ambiguity. If \(\langle\cdot,\cdot\rangle\) is linear in the first argument, then \(v\mapsto\langle x,v\rangle\) is antilinear; if the convention is reversed, the other slot is. The draft describes the dependency rather than declaring one notation universal.

Manages Complexity¶

Antilinearity prevents repeated coordinate calculations involving conjugated scalars from looking like exceptions to linear algebra. By naming the twist, one retains additivity, basis descriptions, kernels, images, rank, continuity, and composition while tracking exactly where conjugation enters.

The conjugate-space equivalence is especially compressive. Instead of rebuilding every construction, an antilinear map can be treated as linear with one endpoint conjugated. Tensor products, duals, operator spaces, and representations can then use ordinary linear machinery as long as the scalar structure is not forgotten.

Composition parity is a second compression. Two conjugations cancel: if (A) and (B) are antilinear, then \(B(A(\lambda v))=B(\overline\lambda A(v))=\lambda B(A(v))\), so \(B\circ A\) is linear. This explains why squares of antiunitary symmetries are linear and can equal (+I), (-I), or another unitary operator.

In computation, the decomposition of a real-linear map (T) into complex-linear and antilinear parts isolates failure of complex linearity:

\[ T_{\mathrm{lin}}=\tfrac12(T-JTJ),\qquad T_{\mathrm{anti}}=\tfrac12(T+JTJ), \]

with (Jv=iv). This makes mixed complex-variable models analyzable without incorrectly forcing them into one class.

Abstract Reasoning¶

If (A) is both complex-linear and antilinear, then (A(iv)=iA(v)=-iA(v)), so (A=0) over characteristic zero.
If (A) and (B) are antilinear, \(B\circ A\) is complex-linear because conjugation is applied twice.
If (L) is linear and (A) antilinear, both \(L\circ A\) and \(A\circ L\) are antilinear.
If a bijective antilinear map has an inverse, the inverse is antilinear.
If a map is antilinear, its kernel is a complex subspace: \(A(\lambda v)=\overline\lambda A(v)=0\).
If \(A(z)=M\overline z\), then \(A^2(z)=M\overline M z\); involutivity requires \(M\overline M=I\), not merely (M^2=I).
If (A) is antiunitary, it is norm preserving and bijective, but it need not square to identity.
If the scalars are restricted to real numbers, the conjugate-homogeneity test cannot distinguish antilinear from linear.
If coordinatewise conjugation changes under a basis change, the chosen conjugation encodes extra real-structure data rather than a basis-free operation.
If a theorem is stated using inner products, changing which slot is linear can reverse whether the associated functional or Riesz map is linear or antilinear without changing the underlying mathematics.
If a real-linear map commutes with (J), it is complex-linear; if it anticommutes with (J), it is antilinear.

Knowledge Transfer¶

The exact abstraction transfers among finite-dimensional vector spaces, complex Banach and Hilbert spaces, operator spaces, complex representations, bundles, and quantum state spaces whenever the same conjugate-scalar law holds. Topology and domains add requirements—boundedness, continuity, closedness, dense definition—but do not alter the algebraic invariant.

It transfers to fields with other automorphisms only through the broader term semilinear. Frobenius-semilinear maps over finite fields, for example, share the twist pattern but are not normally called complex-antilinear. Keeping the automorphism explicit prevents a false alias.

Applications to time reversal, real structures, and conjugate representations transfer the formal map but add their own constraints. Time reversal needs antiunitarity; a real structure needs an involution; a conjugation needs antilinear isometry plus involution. Importing “antilinear” alone does not carry these stronger identities.

Examples¶

complex conjugation: \(K:\mathbb C^n\to\mathbb C^n\), \(K(z)=\overline z\), is an antilinear involutive isometry in the standard basis;
matrix followed by conjugation: \(A(z)=M\overline z\) is the general finite-dimensional coordinate form;
antilinear functional: on a Hilbert space with inner product linear in the first slot, \(v\mapsto\langle x,v\rangle\) is antilinear;
adjoint operation on operator space: \(T\mapsto T^*\) satisfies \((\lambda T)^*=\overline\lambda T^*\), even though each (T^*) acts linearly on vectors;
antiunitary time reversal: an antilinear norm-preserving symmetry conjugates probability amplitudes and reverses the complex structure used in time evolution;
real structure: an antilinear involution on a complex vector space has a real fixed-point space under suitable conditions;
two-factor composition: \(K\circ K=I\) is linear, illustrating parity;
non-example—transpose: \(M\mapsto M^{\mathsf T}\) is complex-linear on matrix space, while conjugate transpose is conjugate-linear;
non-example—absolute value: \(z\mapsto|z|\) is neither additive nor a map of complex vector spaces with conjugate homogeneity.

Structural Tensions¶

linear machinery vs. conjugated scalars — most constructions survive, while one missed bar invalidates scalar reasoning;
coordinate simplicity vs. basis dependence — \(M\overline z\) is concrete, while coordinatewise conjugation encodes a chosen real structure;
algebraic identity vs. analytic properties — antilinearity alone is simple, while boundedness, domains, adjoints, and spectra need separate hypotheses;
general class vs. strong symmetry — antiunitary and conjugation operators are memorable examples, while neither extra condition belongs to every antilinear map;
mathematical vs. physics convention — inner-product slot conventions differ, making unqualified claims about linearity of associated functionals hazardous;
real restriction vs. complex distinction — the classes coincide over real scalars, while their difference controls complex phases and symmetries;
basis-free equivalence vs. implementation — conjugate spaces give invariant meaning, while numerical work must choose coordinates and storage conventions.

Structural–Framed Character¶

Antilinear Map is structural. Once vector spaces and the complex-conjugation automorphism are fixed, the defining equations have determinate truth values. Composition parity, kernel structure, matrix form, and conjugate-space equivalence are deductions, not matters of interpretation.

Conventions affect notation, not identity. Choosing which inner-product slot is linear or which basis supplies coordinate conjugation changes formulas and associated identifications. Stating those choices keeps the formal structure invariant.

Structural Core vs. Domain Accent¶

The structural core is additive operation preserved + scalar action transported through a fixed automorphism + ordinary structure recovered after twisting one endpoint + parity under composition. The broader mathematical home of that core is semilinearity.

The domain accent is complex vector spaces, complex conjugation, conjugate Hilbert spaces, anti-duals, antiunitary symmetries, and real structures. Remove conjugation and one has generic semilinearity or homomorphism. Retain it and the node remains a precise complex-linear-algebra abstraction.

Linearity — addition and a disciplined scalar law remain central, but the scalar is conjugated; this is the smallest prospective parent by presupposition, not subtype.
Transformation — the map converts vectors while preserving specified algebraic relations.
Symmetry — antiunitary and involutive special cases implement conjugating symmetries.
Duality — anti-duals and Riesz representation connect vectors and conjugate-linear functionals.
Involution — conjugation operators and real structures add (C^2=I), though generic antilinear maps do not.
Composition — parity of conjugation count determines whether a composite is linear or antilinear.
Change of Representation — conjugate vector spaces turn antilinear maps into linear maps by changing scalar structure.
Constraint — the condition (AJ=-JA) sharply selects the antilinear component among real-linear maps.

Relationships to Other Abstractions¶

Current abstraction Antilinear Map Domain-specific

Parents (1) — more general patterns this builds on

Antilinear Map presupposes Linearity Prime

addition and a disciplined scalar law remain central, but the scalar is conjugated; this is the smallest prospective parent by presupposition, not subtype.

Hierarchy path (1) — routes to 1 parentless root

Antilinear Map → Linearity

Neighborhood in Abstraction Space¶

Antilinear Map sits in a sparse region of the domain-specific corpus (88^th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.

Family — Unclustered & Miscellaneous (1565 abstractions)

Nearest neighbors

Computed from structural-signature embeddings · 2026-09-08

Not to Be Confused With¶

complex-linear or generic real-linear maps;
arbitrary nonlinear maps;
semilinear maps for automorphisms other than complex conjugation;
antiunitary operators;
conjugations or real structures requiring involution;
adjoints of operators;
entrywise matrix conjugation or conjugate transpose;
antiholomorphic maps in general;
conjugate representations as entire representation-theoretic objects;
the accepted Multilinear Form node, which is linear separately in multiple arguments rather than conjugate-linear in one argument.

References¶

[1] Walter Rudin, Functional Analysis, 2^nd ed. (McGraw–Hill, 1991), chapters on complex topological vector spaces and conjugate-linear functionals. registry ↩a ↩b

[2] S. Hassi and H. S. V. de Snoo, “Antilinear Normal Operators on Hilbert Space,” Integral Equations and Operator Theory (2026), https://doi.org/10.1007/s00020-026-02862-w. registry ↩a ↩b ↩c

[3] Terence Tao, An Epsilon of Room, I: Real Analysis (American Mathematical Society, 2010), discussion of conjugate Hilbert spaces and antilinear isometries, https://www.ams.org/bookstore/pspdf/gsm-117-prev.pdf. registry ↩a ↩b

[4] Eugene P. Wigner, Group Theory and Its Application to the Quantum Mechanics of Atomic Spectra, trans. J. J. Griffin (Academic Press, 1959), discussion of unitary and antiunitary symmetry transformations. registry ↩a ↩b

[5] Brian C. Hall, Quantum Theory for Mathematicians (Springer, 2013), https://doi.org/10.1007/978-1-4614-7116-5. registry ↩a ↩b

[6] “Efficient Iterative Solutions to Complex-Valued Nonlinear Least-Squares Problems with Mixed Linear and Antilinear Operators,” Optimization and Engineering 23 (2022), https://doi.org/10.1007/s11081-021-09604-4. registry

[7] “Antilinear map,” Wikipedia, frozen revision 1363358412, https://en.wikipedia.org/wiki/Antilinear_map. registry