<?xml version="1.0"?>
<feed xmlns="http://www.w3.org/2005/Atom" xml:lang="en">
	<id>https://wiki.sarg.dev/api.php?action=feedcontributions&amp;feedformat=atom&amp;user=2A02%3A2455%3A18A6%3AE200%3A0%3A0%3A0%3ABD61</id>
	<title>Vero - Wikipedia - User contributions [en]</title>
	<link rel="self" type="application/atom+xml" href="https://wiki.sarg.dev/api.php?action=feedcontributions&amp;feedformat=atom&amp;user=2A02%3A2455%3A18A6%3AE200%3A0%3A0%3A0%3ABD61"/>
	<link rel="alternate" type="text/html" href="https://wiki.sarg.dev/index.php/Special:Contributions/2A02:2455:18A6:E200:0:0:0:BD61"/>
	<updated>2026-08-14T16:40:03Z</updated>
	<subtitle>User contributions</subtitle>
	<generator>MediaWiki 1.44.2</generator>
	<entry>
		<id>https://wiki.sarg.dev/index.php?title=Splitting_lemma&amp;diff=101287</id>
		<title>Splitting lemma</title>
		<link rel="alternate" type="text/html" href="https://wiki.sarg.dev/index.php?title=Splitting_lemma&amp;diff=101287"/>
		<updated>2025-01-27T09:22:42Z</updated>

		<summary type="html">&lt;p&gt;2A02:2455:18A6:E200:0:0:0:BD61: rearrange sentence for clarity&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;{{Short description|About direct sums and exact sequences}}&lt;br /&gt;
{{Distinguish|text=the [[splitting lemma (functions)|splitting lemma]] in [[singularity theory]]}}&lt;br /&gt;
&lt;br /&gt;
In [[mathematics]], and more specifically in [[homological algebra]], the &#039;&#039;&#039;splitting lemma&#039;&#039;&#039; states that in any [[abelian category]], the following statements are [[logical equivalence|equivalent]] for a [[short exact sequence]] &lt;br /&gt;
: &amp;lt;math&amp;gt;0 \longrightarrow A \mathrel{\overset{q}{\longrightarrow}} B \mathrel{\overset{r}{\longrightarrow}} C \longrightarrow 0.&amp;lt;/math&amp;gt;&lt;br /&gt;
{{ordered list|{{glossary}}{{term|Left split}}{{defn|There exists a [[morphism]] {{math|&#039;&#039;t&#039;&#039;: &#039;&#039;B&#039;&#039; → &#039;&#039;A&#039;&#039;}} such that {{math|&#039;&#039;tq&#039;&#039;}} is the [[identity function|identity]] {{math|id{{sub|&#039;&#039;A&#039;&#039;}}}} on {{math|&#039;&#039;A&#039;&#039;}},}}{{glossary end}}|{{glossary}}{{term|Right split}}{{defn|There exists a morphism {{math|&#039;&#039;u&#039;&#039;: &#039;&#039;C&#039;&#039; → &#039;&#039;B&#039;&#039;}} such that {{math|&#039;&#039;ru&#039;&#039;}} is the identity {{math|id{{sub|&#039;&#039;C&#039;&#039;}}}} on {{math|&#039;&#039;C&#039;&#039;}},}}{{glossary end}}|{{glossary}}{{term|Direct sum}}{{defn|There is an [[isomorphism (category theory)|isomorphism]] {{mvar|h}} from {{math|&#039;&#039;B&#039;&#039;}} to the [[biproduct|direct sum]] of {{math|&#039;&#039;A&#039;&#039;}} and {{math|&#039;&#039;C&#039;&#039;}}, such that {{math|&#039;&#039;hq&#039;&#039;}} is the natural injection of {{math|&#039;&#039;A&#039;&#039;}} into the direct sum, and &amp;lt;math&amp;gt;rh^{-1}&amp;lt;/math&amp;gt; is the natural projection of the direct sum [[surjective|onto]] {{math|&#039;&#039;C&#039;&#039;}}.}}{{glossary end}}&lt;br /&gt;
}}&lt;br /&gt;
&lt;br /&gt;
If any of these statements holds, the sequence is called a &#039;&#039;&#039;[[split exact sequence]]&#039;&#039;&#039;, and the sequence is said to &#039;&#039;split&#039;&#039;.&lt;br /&gt;
&lt;br /&gt;
In the above short exact sequence, where the sequence splits, it allows one to refine the [[first isomorphism theorem]], which states that:&lt;br /&gt;
: {{math|&#039;&#039;C&#039;&#039; ≅ &#039;&#039;B&#039;&#039;/ker &#039;&#039;r&#039;&#039; ≅ &#039;&#039;B&#039;&#039;/&#039;&#039;q&#039;&#039;(&#039;&#039;A&#039;&#039;)}} (i.e., {{math|&#039;&#039;C&#039;&#039;}} isomorphic to the [[coimage]] of {{math|&#039;&#039;r&#039;&#039;}} or [[cokernel]] of {{math|&#039;&#039;q&#039;&#039;}})&lt;br /&gt;
to:&lt;br /&gt;
: {{math|&#039;&#039;B&#039;&#039; {{=}} &#039;&#039;q&#039;&#039;(&#039;&#039;A&#039;&#039;) ⊕ &#039;&#039;u&#039;&#039;(&#039;&#039;C&#039;&#039;) ≅ &#039;&#039;A&#039;&#039; ⊕ &#039;&#039;C&#039;&#039;}}&lt;br /&gt;
where the first isomorphism theorem is then just the projection onto {{math|&#039;&#039;C&#039;&#039;}}.&lt;br /&gt;
&lt;br /&gt;
It is a [[category theory|categorical]] generalization of the [[rank–nullity theorem]] (in the form {{math|V ≅ ker&amp;amp;thinsp;&#039;&#039;T&#039;&#039; ⊕ im&amp;amp;thinsp;&#039;&#039;T&#039;&#039;)}} in [[linear algebra]].&lt;br /&gt;
&lt;br /&gt;
==Proof for the category of abelian groups==&lt;br /&gt;
&lt;br /&gt;
=== {{math|3. ⇒ 1.}} and {{math|3. ⇒ 2.}} ===&lt;br /&gt;
First, to show that 3. implies both 1. and 2., we assume 3. and take as {{math|&#039;&#039;t&#039;&#039;}} the natural projection of the direct sum onto {{math|&#039;&#039;A&#039;&#039;}}, and take as {{math|&#039;&#039;u&#039;&#039;}} the natural injection of {{math|&#039;&#039;C&#039;&#039;}} into the direct sum.&lt;br /&gt;
&lt;br /&gt;
=== {{math|1. ⇒ 3.}} ===&lt;br /&gt;
To [[mathematical proof|prove]] that 1. implies 3., first note that any member of &#039;&#039;B&#039;&#039; is in the set ({{math|[[kernel (algebra)|ker]] &#039;&#039;t&#039;&#039; + [[image (function)|im]] &#039;&#039;q&#039;&#039;}}). This follows since for all {{math|&#039;&#039;b&#039;&#039;}} in {{math|&#039;&#039;B&#039;&#039;}}, {{math|&#039;&#039;b&#039;&#039; {{=}} (&#039;&#039;b&#039;&#039; − &#039;&#039;qt&#039;&#039;(&#039;&#039;b&#039;&#039;)) + &#039;&#039;qt&#039;&#039;(&#039;&#039;b&#039;&#039;)}}; {{math|&#039;&#039;qt&#039;&#039;(&#039;&#039;b&#039;&#039;)}} is in {{math|im &#039;&#039;q&#039;&#039;}}, and {{math|&#039;&#039;b&#039;&#039; − &#039;&#039;qt&#039;&#039;(&#039;&#039;b&#039;&#039;)}} is in {{math|ker &#039;&#039;t&#039;&#039;}}, since &lt;br /&gt;
:{{math|&#039;&#039;t&#039;&#039;(&#039;&#039;b&#039;&#039; − &#039;&#039;qt&#039;&#039;(&#039;&#039;b&#039;&#039;)) {{=}} &#039;&#039;t&#039;&#039;(&#039;&#039;b&#039;&#039;) − &#039;&#039;tqt&#039;&#039;(&#039;&#039;b&#039;&#039;) {{=}} &#039;&#039;t&#039;&#039;(&#039;&#039;b&#039;&#039;) − (&#039;&#039;tq&#039;&#039;)&#039;&#039;t&#039;&#039;(&#039;&#039;b&#039;&#039;) {{=}} &#039;&#039;t&#039;&#039;(&#039;&#039;b&#039;&#039;) − &#039;&#039;t&#039;&#039;(&#039;&#039;b&#039;&#039;) {{=}} 0.}}&lt;br /&gt;
&lt;br /&gt;
Next, the [[intersection (set theory)|intersection]] of {{math|im &#039;&#039;q&#039;&#039;}} and {{math|ker &#039;&#039;t&#039;&#039;}} is 0, since if there exists {{math|&#039;&#039;a&#039;&#039;}} in {{math|&#039;&#039;A&#039;&#039;}} such that {{math|&#039;&#039;q&#039;&#039;(&#039;&#039;a&#039;&#039;) {{=}} &#039;&#039;b&#039;&#039;}}, and {{math|&#039;&#039;t&#039;&#039;(&#039;&#039;b&#039;&#039;) {{=}} 0}}, then {{math|0 {{=}} &#039;&#039;tq&#039;&#039;(&#039;&#039;a&#039;&#039;) {{=}} &#039;&#039;a&#039;&#039;}}; and therefore, {{math|&#039;&#039;b&#039;&#039; {{=}} 0}}.&lt;br /&gt;
&lt;br /&gt;
This proves that {{math|&#039;&#039;B&#039;&#039;}} is the direct sum of {{math|im &#039;&#039;q&#039;&#039;}} and {{math|ker &#039;&#039;t&#039;&#039;}}. So, for all {{math|&#039;&#039;b&#039;&#039;}} in {{math|&#039;&#039;B&#039;&#039;}}, {{math|&#039;&#039;b&#039;&#039;}} can be uniquely identified by some {{math|&#039;&#039;a&#039;&#039;}} in {{math|&#039;&#039;A&#039;&#039;}}, {{math|&#039;&#039;k&#039;&#039;}} in {{math|ker &#039;&#039;t&#039;&#039;}}, such that {{math|&#039;&#039;b&#039;&#039; {{=}} &#039;&#039;q&#039;&#039;(&#039;&#039;a&#039;&#039;) + &#039;&#039;k&#039;&#039;}}.&lt;br /&gt;
&lt;br /&gt;
By exactness {{math|ker &#039;&#039;r&#039;&#039; {{=}} im &#039;&#039;q&#039;&#039;}}. The subsequence {{math|&#039;&#039;B&#039;&#039; ⟶ &#039;&#039;C&#039;&#039; ⟶ 0}} implies that {{math|&#039;&#039;r&#039;&#039;}} is [[surjective|onto]]; therefore for any {{math|&#039;&#039;c&#039;&#039;}} in {{math|&#039;&#039;C&#039;&#039;}} there exists some {{math|&#039;&#039;b&#039;&#039; {{=}} &#039;&#039;q&#039;&#039;(&#039;&#039;a&#039;&#039;) + &#039;&#039;k&#039;&#039;}} such that {{math|&#039;&#039;c&#039;&#039; {{=}} &#039;&#039;r&#039;&#039;(&#039;&#039;b&#039;&#039;) {{=}} &#039;&#039;r&#039;&#039;(&#039;&#039;q&#039;&#039;(&#039;&#039;a&#039;&#039;) + &#039;&#039;k&#039;&#039;) {{=}} &#039;&#039;r&#039;&#039;(&#039;&#039;k&#039;&#039;)}}. Therefore, for any &#039;&#039;c&#039;&#039; in &#039;&#039;C&#039;&#039;, exists &#039;&#039;k&#039;&#039; in ker &#039;&#039;t&#039;&#039; such that &#039;&#039;c&#039;&#039; = &#039;&#039;r&#039;&#039;(&#039;&#039;k&#039;&#039;), and &#039;&#039;r&#039;&#039;(ker &#039;&#039;t&#039;&#039;) = &#039;&#039;C&#039;&#039;.&lt;br /&gt;
&lt;br /&gt;
If {{math|&#039;&#039;r&#039;&#039;(&#039;&#039;k&#039;&#039;) {{=}} 0}}, then {{math|&#039;&#039;k&#039;&#039;}} is in {{math|im &#039;&#039;q&#039;&#039;}}; since the intersection of {{math|im &#039;&#039;q&#039;&#039;}} and {{math|ker &#039;&#039;t&#039;&#039; {{=}} 0}}, then {{math|&#039;&#039;k&#039;&#039; {{=}} 0}}. Therefore, the [[restriction (mathematics)|restriction]] {{math|&#039;&#039;r&#039;&#039;: ker &#039;&#039;t&#039;&#039; → &#039;&#039;C&#039;&#039;}} is an isomorphism; and {{math|ker &#039;&#039;t&#039;&#039;}} is isomorphic to {{math|&#039;&#039;C&#039;&#039;}}.&lt;br /&gt;
&lt;br /&gt;
Finally, {{math|im &#039;&#039;q&#039;&#039;}} is isomorphic to {{math|&#039;&#039;A&#039;&#039;}} due to the exactness of {{math|0 ⟶ &#039;&#039;A&#039;&#039; ⟶ &#039;&#039;B&#039;&#039;}}; so &#039;&#039;B&#039;&#039; is isomorphic to the direct sum of {{math|&#039;&#039;A&#039;&#039;}} and {{math|&#039;&#039;C&#039;&#039;}}, which proves (3).&lt;br /&gt;
&lt;br /&gt;
=== {{math|2. ⇒ 3.}} ===&lt;br /&gt;
To show that 2. implies 3., we follow a similar argument. Any member of {{math|&#039;&#039;B&#039;&#039;}} is in the set {{math|ker &#039;&#039;r&#039;&#039; + im &#039;&#039;u&#039;&#039;}}; since for all {{math|&#039;&#039;b&#039;&#039;}} in {{math|&#039;&#039;B&#039;&#039;}}, {{math|&#039;&#039;b&#039;&#039; {{=}} (&#039;&#039;b&#039;&#039; − &#039;&#039;ur&#039;&#039;(&#039;&#039;b&#039;&#039;)) + &#039;&#039;ur&#039;&#039;(&#039;&#039;b&#039;&#039;)}}, which is in {{math|ker &#039;&#039;r&#039;&#039; + im &#039;&#039;u&#039;&#039;}}. The intersection of {{math|ker &#039;&#039;r&#039;&#039;}} and {{math|im &#039;&#039;u&#039;&#039;}} is {{math|0}}, since if {{math|&#039;&#039;r&#039;&#039;(&#039;&#039;b&#039;&#039;) {{=}} 0}} and {{math|&#039;&#039;u&#039;&#039;(&#039;&#039;c&#039;&#039;) {{=}} &#039;&#039;b&#039;&#039;}}, then {{math|0 {{=}} &#039;&#039;ru&#039;&#039;(&#039;&#039;c&#039;&#039;) {{=}} &#039;&#039;c&#039;&#039;}}.&lt;br /&gt;
&lt;br /&gt;
By exactness, {{math|im &#039;&#039;q&#039;&#039; {{=}} ker &#039;&#039;r&#039;&#039;}}, and since {{math|&#039;&#039;q&#039;&#039;}} is an [[injective|injection]], {{math|im &#039;&#039;q&#039;&#039;}} is isomorphic to {{math|&#039;&#039;A&#039;&#039;}}, so {{math|&#039;&#039;A&#039;&#039;}} is isomorphic to {{math|ker &#039;&#039;r&#039;&#039;}}. Since {{math|&#039;&#039;ru&#039;&#039;}} is a [[bijection]], {{math|&#039;&#039;u&#039;&#039;}} is an injection, and thus {{math|im &#039;&#039;u&#039;&#039;}} is isomorphic to {{math|&#039;&#039;C&#039;&#039;}}. So {{math|&#039;&#039;B&#039;&#039;}} is again the direct sum of {{math|&#039;&#039;A&#039;&#039;}} and {{math|&#039;&#039;C&#039;&#039;}}.&lt;br /&gt;
&lt;br /&gt;
An alternative &amp;quot;[[abstract nonsense]]&amp;quot; [https://math.stackexchange.com/q/753182 proof of the splitting lemma] may be formulated entirely in [[category theory|category theoretic]] terms.&lt;br /&gt;
&lt;br /&gt;
==Non-abelian groups==&lt;br /&gt;
In the form stated here, the splitting lemma does not hold in the full [[category of groups]], which is not an abelian category.&lt;br /&gt;
&lt;br /&gt;
===Partially true===&lt;br /&gt;
It is partially true: if a short exact sequence of groups is left split or a direct sum (1. or 3.), then all of the conditions hold. For a direct sum this is clear, as one can inject from or project to the summands. For a left split sequence, the map {{math|&#039;&#039;t&#039;&#039; × &#039;&#039;r&#039;&#039;: &#039;&#039;B&#039;&#039; → &#039;&#039;A&#039;&#039; × &#039;&#039;C&#039;&#039;}} gives an isomorphism, so {{math|&#039;&#039;B&#039;&#039;}} is a direct sum (3.), and thus inverting the isomorphism and composing with the natural injection {{math|&#039;&#039;C&#039;&#039; → &#039;&#039;A&#039;&#039; × &#039;&#039;C&#039;&#039;}} gives an injection {{math|&#039;&#039;C&#039;&#039; → &#039;&#039;B&#039;&#039;}} splitting {{math|&#039;&#039;r&#039;&#039;}} (2.).&lt;br /&gt;
&lt;br /&gt;
However, if a short exact sequence of groups is right split (2.), then it need not be left split or a direct sum (neither 1. nor 3. follows): the problem is that the image of the right splitting need not be [[normal subgroup|normal]]. What is true in this case is that {{math|&#039;&#039;B&#039;&#039;}} is a [[semidirect product]], though not in general a [[direct product of groups|direct product]].&lt;br /&gt;
&lt;br /&gt;
===Counterexample===&lt;br /&gt;
To form a counterexample, take the smallest [[non-abelian group]] {{math|&#039;&#039;B&#039;&#039; ≅ &#039;&#039;S&#039;&#039;{{sub|3}}}}, the [[symmetric group]] on three letters.  Let {{math|&#039;&#039;A&#039;&#039;}} denote the [[alternating group|alternating subgroup]], and let {{math|&#039;&#039;C&#039;&#039; {{=}} &#039;&#039;B&#039;&#039;/&#039;&#039;A&#039;&#039; ≅ {±1}}}.  Let {{math|&#039;&#039;q&#039;&#039;}} and {{math|&#039;&#039;r&#039;&#039;}} denote the inclusion map and the [[parity of a permutation|sign]] map respectively, so that&lt;br /&gt;
&lt;br /&gt;
: &amp;lt;math&amp;gt;0 \longrightarrow A \mathrel{\stackrel{q}{\longrightarrow}} B \mathrel{\stackrel{r}{\longrightarrow}} C \longrightarrow 0 &amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
is a short exact sequence.  3. fails, because {{math|&#039;&#039;S&#039;&#039;{{sub|3}}}} is not abelian, but 2. holds: we may define {{math|&#039;&#039;u&#039;&#039;: &#039;&#039;C&#039;&#039; → &#039;&#039;B&#039;&#039;}} by mapping the generator to any [[cyclic permutation|two-cycle]].  Note for completeness that 1. fails: any map {{math|&#039;&#039;t&#039;&#039;: &#039;&#039;B&#039;&#039; → &#039;&#039;A&#039;&#039;}} must map every two-cycle to the [[identity permutation|identity]] because the map has to be a [[group homomorphism]], while the [[order (group theory)|order]] of a two-cycle is 2 which can not be divided by the order of the elements in &#039;&#039;A&#039;&#039; other than the identity element, which is 3 as {{math|&#039;&#039;A&#039;&#039;}} is the alternating subgroup of {{math|&#039;&#039;S&#039;&#039;{{sub|3}}}}, or namely the [[cyclic group]] of [[order of a group|order]] 3. But every [[permutation]] is a product of two-cycles, so {{math|&#039;&#039;t&#039;&#039;}} is the trivial map, whence {{math|&#039;&#039;tq&#039;&#039;: &#039;&#039;A&#039;&#039; → &#039;&#039;A&#039;&#039;}} is the trivial map, not the identity.&lt;br /&gt;
&lt;br /&gt;
==References==&lt;br /&gt;
* [[Saunders Mac Lane]]: &#039;&#039;Homology&#039;&#039;. Reprint of the 1975 edition, Springer Classics in Mathematics, {{ISBN|3-540-58662-8}}, p.&amp;amp;nbsp;16&lt;br /&gt;
* [[Allen Hatcher]]: &#039;&#039;Algebraic Topology&#039;&#039;. 2002, Cambridge University Press, {{ISBN|0-521-79540-0}}, p.&amp;amp;nbsp;147&lt;br /&gt;
&lt;br /&gt;
{{DEFAULTSORT:Splitting Lemma}}&lt;br /&gt;
[[Category:Homological algebra]]&lt;br /&gt;
[[Category:Lemmas in category theory]]&lt;br /&gt;
[[Category:Articles containing proofs]]&lt;/div&gt;</summary>
		<author><name>2A02:2455:18A6:E200:0:0:0:BD61</name></author>
	</entry>
</feed>