1 Definitions and Basic Properties

Let \(R\) be a ring and \(M\) a (left) \(R\)-module.

A module \(S\neq 0\) is simple if it has no nontrivial submodules, equivalently the only submodules of \(S\) are \(0\) and \(S\). A module \(M\) is semisimple if it is a direct sum of simple submodules. In this case, \(M\) is also described as completely reducible or fully reducible, reflecting that its internal structure breaks into irreducible pieces without extensions between them.

A related notion is indecomposable: a module that is not a nontrivial direct sum. Semisimple modules decompose into indecomposable simples, but the category of semisimple objects is substantially more rigid than that of general modules.

1.2 Equivalent characterizations of semisimple modules

Several conditions are equivalent for a module \(M\) over a ring \(R\):

  1. \(M\) is a direct sum of simple submodules.
  2. Every submodule \(N\le M\) is a direct summand of \(M\), meaning \(M\cong N\oplus N'\) for some \(R\)-module \(N'\).
  3. Every short exact sequence \(0\to N\to E\to M\to 0\) with \(N\) semisimple and \(M\) semisimple splits, or more specifically: whenever \(0\to A\to B\to C\to 0\) has \(C\) semisimple, the sequence splits, yielding \(B\cong A\oplus C\).
  4. One may also phrase semisimplicity via projectivity/injectivity properties: semisimple modules are both semisimple projective and semisimple injective in appropriate senses, leading to vanishing of certain extension groups (captured in later sections).

In many texts, the “every submodule is a direct summand” criterion is taken as the definition because it is robust and module-theoretic.

1.3 Direct sum decompositions and uniqueness aspects

If a semisimple module is written as a direct sum of simple modules, the decomposition exhibits a controlled uniqueness:

  • While the specific placement of simple summands can vary, the multiset of isomorphism classes of simple modules occurring in the decomposition is invariant up to permutation, provided the module is expressed as a finite direct sum (and more generally under suitable finiteness or appropriate categorical hypotheses).
  • For arbitrary (possibly infinite) decompositions, one still expects an invariance of multiplicities in a refined sense, though the formulation depends on how one treats infinite direct sums and how multiplicity is defined.

A key point is that semisimplicity prevents “hidden” extension data: there is no nontrivial way to glue simple constituents together. As a result, classification reduces to tracking which simple types occur and with what multiplicity.

1.4 Examples and non-examples

Examples:

  • Over a field \(k\), every \(k\)-vector space is semisimple. Every subspace has a complementary subspace (choose a basis extension), so the module decomposes into one-dimensional simple pieces.
  • Over a semisimple ring (defined in Section 4), every module is semisimple. This includes, for instance, modules over matrix algebras over division rings (in the finite-dimensional setting).

Non-examples:

  • Over \(R=\mathbb{Z}\), the module \(M=\mathbb{Z}/p^n\mathbb{Z}\) is not semisimple for \(n>1\), since it has a proper nontrivial submodule \(p(\mathbb{Z}/p^n)\) that does not split off as a direct summand.
  • More generally, modules with nontrivial extensions between simple factors fail the splitting criterion and thus are not semisimple.

2 Semisimplicity in Terms of Submodules

2.1 Submodules as direct summands

The defining “fully reducible” behavior manifests in the submodule lattice. If \(M\) is semisimple and \(N\le M\), then there exists \(N'\) such that \[ M \cong N \oplus N'. \] Consequently:

  • The quotient \(M/N\) is naturally isomorphic to \(N'\).
  • The inclusion \(N\hookrightarrow M\) and the projection \(M\to N\) are compatible, in the sense that one can recover \(N\) as a retract of \(M\).

This property is strong: it says semisimple modules have no “obstructions” preventing splitting at the level of submodules.

2.2 Finite-length and complete reducibility (module-theoretic viewpoint)

A module is said to have finite length if it admits a composition series with finitely many factors. For modules of finite length, complete reducibility can be characterized cleanly: being semisimple is equivalent to every submodule being generated by direct summands of simple modules, and equivalently to every short exact sequence with semisimple endpoint splitting.

Thus, within the class of finite-length modules, the terminology “semisimple” aligns closely with “completely reducible,” since composition factors can be separated without extension.

2.3 Behavior under isomorphisms and restriction of scalars

Semisimplicity is invariant under isomorphism: if \(M\cong M'\), then \(M\) is semisimple iff \(M'\) is semisimple, since decomposition into simple summands and the direct-summand property are preserved by isomorphisms.

Under restriction of scalars (viewing an \(R\)-module as a module over a subring \(S\subseteq R\)), semisimplicity may be preserved or may fail depending on how simples for \(R\) relate to simples for \(S\). If the restriction functor preserves simplicity and splitting in the relevant way, semisimplicity carries over; otherwise, additional composition behavior can appear.

2.4 Closure properties: sums, products, and quotients

Semisimple modules enjoy several closure features, though the precise statements depend on whether one allows infinite direct sums/products.

  • Direct sums: A direct sum of semisimple modules is semisimple. Each summand splits all of its submodules, and the resulting submodule structure across the sum remains decomposable into simple parts.
  • Direct products: For many algebraic settings, a direct product of semisimple modules is again semisimple if the definition and category allow the relevant splitting of submodules; however, infinite products can require care, since submodules of a product need not decompose componentwise without additional conditions.
  • Quotients: If \(M\) is semisimple and \(N\le M\), then \(M/N\) is semisimple. This follows because \(N\) is a direct summand and thus the quotient is isomorphic to the complementary summand.

These closure properties are among the practical advantages of the semisimple condition.

2.5 Counterexamples showing limitations

Not every desirable closure property is automatic in arbitrary generality.

  • Infinite products may fail to be semisimple in some contexts, or may require assumptions (such as finite generation, artinian hypotheses, or appropriate finiteness constraints) to ensure that all submodules split.
  • Restriction of scalars can destroy semisimplicity when a module that splits as \(R\)-simples becomes a non-splitting extension as an \(S\)-module.

Such counterexamples highlight that semisimplicity is not purely “internal” to the underlying abelian group—it is tied to the module structure over the chosen ring.

3 Structural Results and Module Operations

3.1 Submodules, quotient modules, and semisimple heredity

A central stability principle is heredity: semisimplicity passes to submodules and quotients.

  • Submodules split off by definition: if \(M\) is semisimple, then for each \(N\le M\), there is a complementary \(N'\) with \(M\cong N\oplus N'\).
  • Quotients inherit semisimplicity: \(M/N\cong N'\), hence decomposes as a direct sum of simple modules.

This “two-way” inheritance (submodules and quotients) is stronger than what holds for many weaker notions such as artinian or noetherian modules.

3.2 Extensions between simple modules

Nontrivial extensions are the mechanism that prevents semisimplicity. If \(S\) and \(T\) are simple modules, then a semisimple module built from them contains no non-split short exact sequence of the form \[ 0\to S \to E \to T \to 0 \] unless \(E\cong S\oplus T\). In other words, semisimplicity forces the category to behave as if simples were “rigid,” with every extension splitting.

This viewpoint links semisimplicity to vanishing of extension groups, elaborated in Section 5.

3.3 Role of split short exact sequences

A short exact sequence \[ 0\to A \to B \to C \to 0 \] is split if \(B\cong A\oplus C\). For semisimple modules, splitting is ubiquitous:

  • If \(C\) is semisimple, then any such sequence with quotient \(C\) splits.
  • Dually, if \(A\) is semisimple, then the sequence also splits when viewed appropriately via injective/projective conditions.

This splitting behavior makes semisimple modules particularly tractable for computations with module homomorphisms and endomorphisms.

3.4 Decomposing endomorphisms of semisimple modules

Endomorphisms of a semisimple module decompose according to its simple summands. Concretely, if \[ M \cong \bigoplus_{i} S_i^{(\kappa_i)} \] (where \(S_i\) are pairwise non-isomorphic simple modules and \(\kappa_i\) are multiplicities), then endomorphisms can be represented blockwise relative to this decomposition. Maps between non-isomorphic simples vanish, while maps between isomorphic simples form matrix-like structures over their endomorphism rings.

This yields a structured description of \(\mathrm{End}_R(M)\) in terms of endomorphisms of the constituent simples, discussed more directly in Section 6.

4 Semisimple Modules Over Rings

4.1 Semisimple rings vs semisimple modules

A semisimple ring is a ring whose regular module is semisimple (equivalently, the ring is semisimple as a module over itself). For such rings, every module is semisimple.

Conversely, even if a ring is not semisimple, certain modules can still be semisimple. Thus:

  • “Semisimple ring” is a global property of the ring.
  • “Semisimple module” is a property relative to a chosen module over the ring.

This distinction helps separate results about all modules from results about particular module classes.

4.2 Conditions on the ring for semisimplicity of modules

Semisimplicity of modules often reflects how the ring’s internal structure controls extensions. When the ring has strong radical or homological constraints, many modules automatically become semisimple.

Common ring conditions that ensure semisimplicity include:

  • vanishing of certain radicals,
  • ensuring that the module category has enough splitting behavior,
  • or enforcing semisimple decomposition theorems for finitely generated modules.

Rather than listing one universal criterion, standard approaches relate semisimplicity of modules to properties of projective/injective dimensions and to radical annihilation.

4.3 Jacobson radical and implications for semisimple modules

The Jacobson radical \(J(R)\) governs how far a ring is from being semisimple. If \(R\) is semisimple, then \(J(R)=0\). For modules, radicals act as obstructions to splitting: when radicals vanish on a module or when the module is annihilated by a relevant radical, semisimplicity frequently follows.

At the module level, one can view semisimplicity as the situation where the module has no “radical layers” that persist under iteration: the module decomposes entirely into minimal building blocks (simples) without leftover nilpotent or extension effects.

4.4 Wedderburn-type decomposition themes (module-level perspective)

Wedderburn’s structural ideas for semisimple rings manifest module-theoretically as decomposition into simple constituents. In a semisimple environment, the module category behaves like a direct sum of representation types determined by simple modules.

This produces patterns such as:

  • endomorphism rings of simples being division rings,
  • endomorphism rings of semisimple modules decomposing into products of matrix-like algebras over division rings,
  • and a “no mixing” phenomenon between non-isomorphic simple types.

Although the explicit Wedderburn decomposition is often phrased for rings, its spirit appears repeatedly in module decompositions and in describing \(\mathrm{End}_R(M)\).

5 Homological and Categorical Perspectives

5.1 Ext and projectivity/injectivity connections

Semisimplicity can be characterized via the vanishing of extension groups. In abelian categories of modules, one has the principle that:

  • A module \(M\) is semisimple precisely when \(\mathrm{Ext}^1_R(S,M)=0\) for all simple modules \(S\) (and similarly in a dual form using \(\mathrm{Ext}^1_R(M,S)\)).

Equivalently, semisimple modules are both projective and injective with respect to short exact sequences whose other end is simple (and, more generally, with respect to sequences in which splitting is forced).

Thus, extension groups serve as a quantitative measure of how far a module is from being fully reducible.

5.2 Semisimplicity as exactness behavior

Another categorical viewpoint is that semisimplicity makes certain functors behave more like exact functors. Because every relevant short exact sequence splits, one obtains:

  • strong control over kernels and cokernels in diagram chases,
  • and the ability to “linearize” many homological computations.

In practice, semisimplicity turns complicated exactness questions into decompositions into direct summands, simplifying calculations in both representation theory and module theory.

5.3 Semisimple objects and splitting in abelian categories

The definition extends beyond modules: in any abelian category, an object is semisimple if it is a direct sum of simple objects. In such a setting, the same splitting phenomena occur:

  • subobjects are summands,
  • exact sequences with semisimple objects split,
  • and composition factors fully determine the object up to isomorphism (subject to standard categorical conditions).

This categorical framing unifies module-theoretic semisimplicity with semisimple objects in more general algebraic contexts.

5.4 Functorial characterizations

Semisimplicity has functorial signatures. For example:

  • Hom functors from a semisimple module turn exact sequences into split behavior on the level of induced maps.
  • Adjunctions and equivalences between module categories can preserve semisimplicity if they reflect direct sum decompositions and splitting.

In categorical terms, semisimple objects behave like “atoms with no extension,” which makes them stable under many constructions that preserve direct sums and kernels/cokernels appropriately.

6 Decomposition into Simple Summands

6.1 Isotypic components and multiplicities

Given a semisimple module \(M\), one can group simple summands by isomorphism type. For a simple module \(S\), the \(S\)-isotypic component collects all submodules isomorphic to direct sums of copies of \(S\). Then \(M\) decomposes as a direct sum over the distinct simple isomorphism classes: \[ M \cong \bigoplus_{[S]} M_{[S]}, \] where each \(M_{[S]}\) is built from copies of \(S\) alone. Multiplicities describe how many copies occur (with the appropriate meaning for the given finiteness setting).

This organization helps compute endomorphisms and interpret homomorphisms between semisimple modules.

6.2 Endomorphism rings of semisimple modules

The endomorphism ring \(\mathrm{End}_R(M)\) can be described via the simple decomposition. A typical pattern is:

  • \(\mathrm{Hom}_R(S,T)=0\) for non-isomorphic simple modules \(S\not\cong T\).
  • \(\mathrm{End}_R(S)\) is a division ring for a simple module \(S\) (a module-theoretic version of Schur-type behavior).
  • For isotypic components \(S^{(\kappa)}\), endomorphisms correspond to matrices over \(\mathrm{End}_R(S)\), with size governed by the multiplicity \(\kappa\).

Hence, endomorphisms of semisimple modules become “block diagonal” with respect to simple types.

6.3 Interactions with Schur’s lemma

Schur’s lemma states that for simple modules \(S\) and \(T\),

  • \(\mathrm{Hom}_R(S,T)=0\) if \(S\not\cong T\),
  • and \(\mathrm{End}_R(S)\) is a division ring.

Semisimplicity amplifies this: since any homomorphism between direct sums of simples decomposes into maps between simple summands, Schur’s lemma implies most off-diagonal homomorphisms vanish. This yields strong structural simplifications for hom spaces and endomorphism rings.

6.4 Practical decomposition strategies in examples

In concrete settings, semisimple decompositions can be found through:

  • identifying simple submodules and verifying that they split off as direct summands,
  • using known classification of simples (e.g., highest-weight theory in certain representation categories, or modules over specific algebras),
  • computing idempotents associated to summands: in semisimple contexts, idempotents correspond to direct sum decompositions.

Because semisimplicity guarantees splitting, once a simple submodule is recognized, the task reduces to extracting it and repeating the process on the complement.

7.1 Completely reducible vs semisimple (terminology and distinctions)

“Completely reducible” and “semisimple” are often used interchangeably in module theory: both refer to decomposability into simple submodules and the splitting of short exact sequences. In some texts, “completely reducible” is reserved for finite-dimensional representations over fields, while “semisimple” is used more broadly, but the underlying structural idea remains the same.

7.2 Artinian/noetherian modules with semisimple structure

If a module is artinian or noetherian and semisimple, its decomposition into simple summands is typically well-behaved with finite multiplicities in appropriate senses. Finiteness conditions ensure that ascending/descending chain behavior terminates, which aligns with the clean layered structure of semisimplicity.

Thus, semisimplicity plus chain conditions often produces stronger classification and more computable invariants.

7.3 Semisimple submodules vs semisimple quotient modules

A module can have semisimple quotients without being semisimple itself, and it can contain semisimple submodules that do not force global splitting. For example:

  • If \(M\) is not semisimple, there may exist a quotient \(M/N\) that is semisimple because the non-splitting behavior lies elsewhere in the module.
  • Conversely, a module may have semisimple submodules while still admitting nontrivial extensions in the complement.

Semisimplicity of the whole module is stronger than semisimplicity of selected submodules or quotients; the direct-summand property must hold uniformly.

7.4 Semisimple vs semiperfect/other nearby notions

Nearby ring and module notions relax the strict splitting requirements:

  • A semiperfect ring generalizes aspects of projective covers and idempotent lifting; modules over semiperfect rings can exhibit more complicated decomposition behavior than purely semisimple ones.
  • Radical-related conditions such as semiprimitive rings or modules with nilpotent radicals sit between semisimplicity and generality.

These variants reflect different compromises: they may allow nontrivial extensions while still enforcing some structural control. Semisimplicity sits at the endpoint where all such extension data disappears.