Does the tensor product preserve coreflexive equalizers?
Clash Royale CLAN TAG#URR8PPP
up vote
2
down vote
favorite
Let $Bbbk$ be any commutative ring and $otimes$ denote the tensor product over $Bbbk$. By a coreflexive pair I mean a parallel pair of $Bbbk$-module homomorphisms admitting a common retraction, i.e. a pair of arrows $f,g:Mto N$ such that there exists $r:Nto M$ satisfying $rcirc f=mathsfid_M=rcirc g$. By a coreflexive equalizer I mean the equalizer of a coreflexive pair.
Q: Does $Potimes -$ preserve coreflexive equalizers for every $Bbbk$-module $P$?
I know that $Potimes -$ does not preserve equalizers in general, but a coreflexive equalizer is something more particular.
abstract-algebra modules tensor-products
add a comment |Â
up vote
2
down vote
favorite
Let $Bbbk$ be any commutative ring and $otimes$ denote the tensor product over $Bbbk$. By a coreflexive pair I mean a parallel pair of $Bbbk$-module homomorphisms admitting a common retraction, i.e. a pair of arrows $f,g:Mto N$ such that there exists $r:Nto M$ satisfying $rcirc f=mathsfid_M=rcirc g$. By a coreflexive equalizer I mean the equalizer of a coreflexive pair.
Q: Does $Potimes -$ preserve coreflexive equalizers for every $Bbbk$-module $P$?
I know that $Potimes -$ does not preserve equalizers in general, but a coreflexive equalizer is something more particular.
abstract-algebra modules tensor-products
add a comment |Â
up vote
2
down vote
favorite
up vote
2
down vote
favorite
Let $Bbbk$ be any commutative ring and $otimes$ denote the tensor product over $Bbbk$. By a coreflexive pair I mean a parallel pair of $Bbbk$-module homomorphisms admitting a common retraction, i.e. a pair of arrows $f,g:Mto N$ such that there exists $r:Nto M$ satisfying $rcirc f=mathsfid_M=rcirc g$. By a coreflexive equalizer I mean the equalizer of a coreflexive pair.
Q: Does $Potimes -$ preserve coreflexive equalizers for every $Bbbk$-module $P$?
I know that $Potimes -$ does not preserve equalizers in general, but a coreflexive equalizer is something more particular.
abstract-algebra modules tensor-products
Let $Bbbk$ be any commutative ring and $otimes$ denote the tensor product over $Bbbk$. By a coreflexive pair I mean a parallel pair of $Bbbk$-module homomorphisms admitting a common retraction, i.e. a pair of arrows $f,g:Mto N$ such that there exists $r:Nto M$ satisfying $rcirc f=mathsfid_M=rcirc g$. By a coreflexive equalizer I mean the equalizer of a coreflexive pair.
Q: Does $Potimes -$ preserve coreflexive equalizers for every $Bbbk$-module $P$?
I know that $Potimes -$ does not preserve equalizers in general, but a coreflexive equalizer is something more particular.
abstract-algebra modules tensor-products
asked Aug 26 at 11:14
Ender Wiggins
755320
755320
add a comment |Â
add a comment |Â
1 Answer
1
active
oldest
votes
up vote
2
down vote
accepted
There is always a natural map $eta : ker(f-g)otimes Pto ker((f-g)otimes P)$, but it is not always an isomorphism, as the following example shows.
Consider $M=N= bigoplus mathbb Z$ indexed over the naturals, and the maps $f(x) = (0,x_1,x_2,ldots)$, $g(x) = (2x_1,x_1,x_2,ldots)$, and $r(x) = (x_2,x_3,ldots)$, so that $rf=rg=1$ and $a = g-f$ is given by $a(x) = (2x_1,0,0,ldots)$. Then $K=ker a = langle (0,x_2,x_3,ldots)rangle$, while $mathbb Z/2otimes a = 0$. It follows that the map $eta$ is not an isomorphism.
Using this idea, you can prove that tensor products preserve equalizers iff they preserve the reflexive ones. Indeed, consider any map of $mathbb k$-modules $f:Mto N$ and let $F : widetildeMto Nopluswidetilde M$ be the map $(m_1,m_2,ldots)mapsto (fm_1,m_1,m_2,ldots)$, where the tilde means taking a sum. Let $G$ be the map where $f=0$. Then $F$ and $G$ admit a common retraction given by sending $(n,m_1,m_2,ldots)$ to $(m_1,m_2,ldots)$, and $K=ker(F-G)$ is $ker foplus widetilde M$.
Then you obtain that the map $Kotimes Pto ker((F-G)otimes P)$ is just the map with coordinates the natural map $ker fotimes Pto ker(fotimes P)$ and the identity of $widetilde Motimes P$, so the claim follows.
I beg your pardon, Pedro, but I cannot see how this allows me to conclude that $Potimes mathsfEq(f,g)cong mathsfEq(Potimes f,Potimes g)$?
â Ender Wiggins
Aug 26 at 13:55
@EnderWiggins My bad, I misread what you were asking. Fixed.
â Pedro Tamaroffâ¦
Aug 26 at 15:01
Ok, I followed your argument and indeed the conclusion is that $ker((F-G)otimes P)cong ker(fotimes P)oplusbigoplus(Motimes P)$, $ker(F-G)otimes Pcong (ker(f)otimes P) oplus(bigoplus Motimes P)$ and so $-otimes P$ preserves equalizers if and only if it preserves coreflexive ones. However, there's a point in the example you chose that I find obscure: $mathbbZ_2otimes ker(a) = mathbbZ_2otimes(0oplus bigoplus mathbbZ) cong bigoplus mathbbZ_2 cong ker(mathbbZ_2otimes a)$.
â Ender Wiggins
Aug 26 at 15:43
What about $M=bigoplus mathbbZ$, $N=mathbbZ_2oplusbigoplusmathbbZ$, $f(x)=([0],x_1,x_2,ldots)$, $g(x)=([x_1],x_1,x_2,ldots)$?In such a case $ker(f-g)=2mathbbZoplusbigoplus mathbbZ$, whence $ker(f-g)otimesmathbbZ_2=(2mathbbZotimes mathbbZ_2)oplusbigoplus mathbbZ_2$ while $ker((f-g)otimesmathbbZ_2)=bigoplus mathbbZ_2$
â Ender Wiggins
Aug 26 at 15:54
1
@EnderWiggins The equalizers may be abstractly isomorphic, but saying that the functor preserves them means the natural map is an isomorphism.
â Pedro Tamaroffâ¦
Aug 26 at 16:16
 |Â
show 1 more comment
1 Answer
1
active
oldest
votes
1 Answer
1
active
oldest
votes
active
oldest
votes
active
oldest
votes
up vote
2
down vote
accepted
There is always a natural map $eta : ker(f-g)otimes Pto ker((f-g)otimes P)$, but it is not always an isomorphism, as the following example shows.
Consider $M=N= bigoplus mathbb Z$ indexed over the naturals, and the maps $f(x) = (0,x_1,x_2,ldots)$, $g(x) = (2x_1,x_1,x_2,ldots)$, and $r(x) = (x_2,x_3,ldots)$, so that $rf=rg=1$ and $a = g-f$ is given by $a(x) = (2x_1,0,0,ldots)$. Then $K=ker a = langle (0,x_2,x_3,ldots)rangle$, while $mathbb Z/2otimes a = 0$. It follows that the map $eta$ is not an isomorphism.
Using this idea, you can prove that tensor products preserve equalizers iff they preserve the reflexive ones. Indeed, consider any map of $mathbb k$-modules $f:Mto N$ and let $F : widetildeMto Nopluswidetilde M$ be the map $(m_1,m_2,ldots)mapsto (fm_1,m_1,m_2,ldots)$, where the tilde means taking a sum. Let $G$ be the map where $f=0$. Then $F$ and $G$ admit a common retraction given by sending $(n,m_1,m_2,ldots)$ to $(m_1,m_2,ldots)$, and $K=ker(F-G)$ is $ker foplus widetilde M$.
Then you obtain that the map $Kotimes Pto ker((F-G)otimes P)$ is just the map with coordinates the natural map $ker fotimes Pto ker(fotimes P)$ and the identity of $widetilde Motimes P$, so the claim follows.
I beg your pardon, Pedro, but I cannot see how this allows me to conclude that $Potimes mathsfEq(f,g)cong mathsfEq(Potimes f,Potimes g)$?
â Ender Wiggins
Aug 26 at 13:55
@EnderWiggins My bad, I misread what you were asking. Fixed.
â Pedro Tamaroffâ¦
Aug 26 at 15:01
Ok, I followed your argument and indeed the conclusion is that $ker((F-G)otimes P)cong ker(fotimes P)oplusbigoplus(Motimes P)$, $ker(F-G)otimes Pcong (ker(f)otimes P) oplus(bigoplus Motimes P)$ and so $-otimes P$ preserves equalizers if and only if it preserves coreflexive ones. However, there's a point in the example you chose that I find obscure: $mathbbZ_2otimes ker(a) = mathbbZ_2otimes(0oplus bigoplus mathbbZ) cong bigoplus mathbbZ_2 cong ker(mathbbZ_2otimes a)$.
â Ender Wiggins
Aug 26 at 15:43
What about $M=bigoplus mathbbZ$, $N=mathbbZ_2oplusbigoplusmathbbZ$, $f(x)=([0],x_1,x_2,ldots)$, $g(x)=([x_1],x_1,x_2,ldots)$?In such a case $ker(f-g)=2mathbbZoplusbigoplus mathbbZ$, whence $ker(f-g)otimesmathbbZ_2=(2mathbbZotimes mathbbZ_2)oplusbigoplus mathbbZ_2$ while $ker((f-g)otimesmathbbZ_2)=bigoplus mathbbZ_2$
â Ender Wiggins
Aug 26 at 15:54
1
@EnderWiggins The equalizers may be abstractly isomorphic, but saying that the functor preserves them means the natural map is an isomorphism.
â Pedro Tamaroffâ¦
Aug 26 at 16:16
 |Â
show 1 more comment
up vote
2
down vote
accepted
There is always a natural map $eta : ker(f-g)otimes Pto ker((f-g)otimes P)$, but it is not always an isomorphism, as the following example shows.
Consider $M=N= bigoplus mathbb Z$ indexed over the naturals, and the maps $f(x) = (0,x_1,x_2,ldots)$, $g(x) = (2x_1,x_1,x_2,ldots)$, and $r(x) = (x_2,x_3,ldots)$, so that $rf=rg=1$ and $a = g-f$ is given by $a(x) = (2x_1,0,0,ldots)$. Then $K=ker a = langle (0,x_2,x_3,ldots)rangle$, while $mathbb Z/2otimes a = 0$. It follows that the map $eta$ is not an isomorphism.
Using this idea, you can prove that tensor products preserve equalizers iff they preserve the reflexive ones. Indeed, consider any map of $mathbb k$-modules $f:Mto N$ and let $F : widetildeMto Nopluswidetilde M$ be the map $(m_1,m_2,ldots)mapsto (fm_1,m_1,m_2,ldots)$, where the tilde means taking a sum. Let $G$ be the map where $f=0$. Then $F$ and $G$ admit a common retraction given by sending $(n,m_1,m_2,ldots)$ to $(m_1,m_2,ldots)$, and $K=ker(F-G)$ is $ker foplus widetilde M$.
Then you obtain that the map $Kotimes Pto ker((F-G)otimes P)$ is just the map with coordinates the natural map $ker fotimes Pto ker(fotimes P)$ and the identity of $widetilde Motimes P$, so the claim follows.
I beg your pardon, Pedro, but I cannot see how this allows me to conclude that $Potimes mathsfEq(f,g)cong mathsfEq(Potimes f,Potimes g)$?
â Ender Wiggins
Aug 26 at 13:55
@EnderWiggins My bad, I misread what you were asking. Fixed.
â Pedro Tamaroffâ¦
Aug 26 at 15:01
Ok, I followed your argument and indeed the conclusion is that $ker((F-G)otimes P)cong ker(fotimes P)oplusbigoplus(Motimes P)$, $ker(F-G)otimes Pcong (ker(f)otimes P) oplus(bigoplus Motimes P)$ and so $-otimes P$ preserves equalizers if and only if it preserves coreflexive ones. However, there's a point in the example you chose that I find obscure: $mathbbZ_2otimes ker(a) = mathbbZ_2otimes(0oplus bigoplus mathbbZ) cong bigoplus mathbbZ_2 cong ker(mathbbZ_2otimes a)$.
â Ender Wiggins
Aug 26 at 15:43
What about $M=bigoplus mathbbZ$, $N=mathbbZ_2oplusbigoplusmathbbZ$, $f(x)=([0],x_1,x_2,ldots)$, $g(x)=([x_1],x_1,x_2,ldots)$?In such a case $ker(f-g)=2mathbbZoplusbigoplus mathbbZ$, whence $ker(f-g)otimesmathbbZ_2=(2mathbbZotimes mathbbZ_2)oplusbigoplus mathbbZ_2$ while $ker((f-g)otimesmathbbZ_2)=bigoplus mathbbZ_2$
â Ender Wiggins
Aug 26 at 15:54
1
@EnderWiggins The equalizers may be abstractly isomorphic, but saying that the functor preserves them means the natural map is an isomorphism.
â Pedro Tamaroffâ¦
Aug 26 at 16:16
 |Â
show 1 more comment
up vote
2
down vote
accepted
up vote
2
down vote
accepted
There is always a natural map $eta : ker(f-g)otimes Pto ker((f-g)otimes P)$, but it is not always an isomorphism, as the following example shows.
Consider $M=N= bigoplus mathbb Z$ indexed over the naturals, and the maps $f(x) = (0,x_1,x_2,ldots)$, $g(x) = (2x_1,x_1,x_2,ldots)$, and $r(x) = (x_2,x_3,ldots)$, so that $rf=rg=1$ and $a = g-f$ is given by $a(x) = (2x_1,0,0,ldots)$. Then $K=ker a = langle (0,x_2,x_3,ldots)rangle$, while $mathbb Z/2otimes a = 0$. It follows that the map $eta$ is not an isomorphism.
Using this idea, you can prove that tensor products preserve equalizers iff they preserve the reflexive ones. Indeed, consider any map of $mathbb k$-modules $f:Mto N$ and let $F : widetildeMto Nopluswidetilde M$ be the map $(m_1,m_2,ldots)mapsto (fm_1,m_1,m_2,ldots)$, where the tilde means taking a sum. Let $G$ be the map where $f=0$. Then $F$ and $G$ admit a common retraction given by sending $(n,m_1,m_2,ldots)$ to $(m_1,m_2,ldots)$, and $K=ker(F-G)$ is $ker foplus widetilde M$.
Then you obtain that the map $Kotimes Pto ker((F-G)otimes P)$ is just the map with coordinates the natural map $ker fotimes Pto ker(fotimes P)$ and the identity of $widetilde Motimes P$, so the claim follows.
There is always a natural map $eta : ker(f-g)otimes Pto ker((f-g)otimes P)$, but it is not always an isomorphism, as the following example shows.
Consider $M=N= bigoplus mathbb Z$ indexed over the naturals, and the maps $f(x) = (0,x_1,x_2,ldots)$, $g(x) = (2x_1,x_1,x_2,ldots)$, and $r(x) = (x_2,x_3,ldots)$, so that $rf=rg=1$ and $a = g-f$ is given by $a(x) = (2x_1,0,0,ldots)$. Then $K=ker a = langle (0,x_2,x_3,ldots)rangle$, while $mathbb Z/2otimes a = 0$. It follows that the map $eta$ is not an isomorphism.
Using this idea, you can prove that tensor products preserve equalizers iff they preserve the reflexive ones. Indeed, consider any map of $mathbb k$-modules $f:Mto N$ and let $F : widetildeMto Nopluswidetilde M$ be the map $(m_1,m_2,ldots)mapsto (fm_1,m_1,m_2,ldots)$, where the tilde means taking a sum. Let $G$ be the map where $f=0$. Then $F$ and $G$ admit a common retraction given by sending $(n,m_1,m_2,ldots)$ to $(m_1,m_2,ldots)$, and $K=ker(F-G)$ is $ker foplus widetilde M$.
Then you obtain that the map $Kotimes Pto ker((F-G)otimes P)$ is just the map with coordinates the natural map $ker fotimes Pto ker(fotimes P)$ and the identity of $widetilde Motimes P$, so the claim follows.
edited Aug 26 at 15:12
answered Aug 26 at 13:13
Pedro Tamaroffâ¦
94.4k10143291
94.4k10143291
I beg your pardon, Pedro, but I cannot see how this allows me to conclude that $Potimes mathsfEq(f,g)cong mathsfEq(Potimes f,Potimes g)$?
â Ender Wiggins
Aug 26 at 13:55
@EnderWiggins My bad, I misread what you were asking. Fixed.
â Pedro Tamaroffâ¦
Aug 26 at 15:01
Ok, I followed your argument and indeed the conclusion is that $ker((F-G)otimes P)cong ker(fotimes P)oplusbigoplus(Motimes P)$, $ker(F-G)otimes Pcong (ker(f)otimes P) oplus(bigoplus Motimes P)$ and so $-otimes P$ preserves equalizers if and only if it preserves coreflexive ones. However, there's a point in the example you chose that I find obscure: $mathbbZ_2otimes ker(a) = mathbbZ_2otimes(0oplus bigoplus mathbbZ) cong bigoplus mathbbZ_2 cong ker(mathbbZ_2otimes a)$.
â Ender Wiggins
Aug 26 at 15:43
What about $M=bigoplus mathbbZ$, $N=mathbbZ_2oplusbigoplusmathbbZ$, $f(x)=([0],x_1,x_2,ldots)$, $g(x)=([x_1],x_1,x_2,ldots)$?In such a case $ker(f-g)=2mathbbZoplusbigoplus mathbbZ$, whence $ker(f-g)otimesmathbbZ_2=(2mathbbZotimes mathbbZ_2)oplusbigoplus mathbbZ_2$ while $ker((f-g)otimesmathbbZ_2)=bigoplus mathbbZ_2$
â Ender Wiggins
Aug 26 at 15:54
1
@EnderWiggins The equalizers may be abstractly isomorphic, but saying that the functor preserves them means the natural map is an isomorphism.
â Pedro Tamaroffâ¦
Aug 26 at 16:16
 |Â
show 1 more comment
I beg your pardon, Pedro, but I cannot see how this allows me to conclude that $Potimes mathsfEq(f,g)cong mathsfEq(Potimes f,Potimes g)$?
â Ender Wiggins
Aug 26 at 13:55
@EnderWiggins My bad, I misread what you were asking. Fixed.
â Pedro Tamaroffâ¦
Aug 26 at 15:01
Ok, I followed your argument and indeed the conclusion is that $ker((F-G)otimes P)cong ker(fotimes P)oplusbigoplus(Motimes P)$, $ker(F-G)otimes Pcong (ker(f)otimes P) oplus(bigoplus Motimes P)$ and so $-otimes P$ preserves equalizers if and only if it preserves coreflexive ones. However, there's a point in the example you chose that I find obscure: $mathbbZ_2otimes ker(a) = mathbbZ_2otimes(0oplus bigoplus mathbbZ) cong bigoplus mathbbZ_2 cong ker(mathbbZ_2otimes a)$.
â Ender Wiggins
Aug 26 at 15:43
What about $M=bigoplus mathbbZ$, $N=mathbbZ_2oplusbigoplusmathbbZ$, $f(x)=([0],x_1,x_2,ldots)$, $g(x)=([x_1],x_1,x_2,ldots)$?In such a case $ker(f-g)=2mathbbZoplusbigoplus mathbbZ$, whence $ker(f-g)otimesmathbbZ_2=(2mathbbZotimes mathbbZ_2)oplusbigoplus mathbbZ_2$ while $ker((f-g)otimesmathbbZ_2)=bigoplus mathbbZ_2$
â Ender Wiggins
Aug 26 at 15:54
1
@EnderWiggins The equalizers may be abstractly isomorphic, but saying that the functor preserves them means the natural map is an isomorphism.
â Pedro Tamaroffâ¦
Aug 26 at 16:16
I beg your pardon, Pedro, but I cannot see how this allows me to conclude that $Potimes mathsfEq(f,g)cong mathsfEq(Potimes f,Potimes g)$?
â Ender Wiggins
Aug 26 at 13:55
I beg your pardon, Pedro, but I cannot see how this allows me to conclude that $Potimes mathsfEq(f,g)cong mathsfEq(Potimes f,Potimes g)$?
â Ender Wiggins
Aug 26 at 13:55
@EnderWiggins My bad, I misread what you were asking. Fixed.
â Pedro Tamaroffâ¦
Aug 26 at 15:01
@EnderWiggins My bad, I misread what you were asking. Fixed.
â Pedro Tamaroffâ¦
Aug 26 at 15:01
Ok, I followed your argument and indeed the conclusion is that $ker((F-G)otimes P)cong ker(fotimes P)oplusbigoplus(Motimes P)$, $ker(F-G)otimes Pcong (ker(f)otimes P) oplus(bigoplus Motimes P)$ and so $-otimes P$ preserves equalizers if and only if it preserves coreflexive ones. However, there's a point in the example you chose that I find obscure: $mathbbZ_2otimes ker(a) = mathbbZ_2otimes(0oplus bigoplus mathbbZ) cong bigoplus mathbbZ_2 cong ker(mathbbZ_2otimes a)$.
â Ender Wiggins
Aug 26 at 15:43
Ok, I followed your argument and indeed the conclusion is that $ker((F-G)otimes P)cong ker(fotimes P)oplusbigoplus(Motimes P)$, $ker(F-G)otimes Pcong (ker(f)otimes P) oplus(bigoplus Motimes P)$ and so $-otimes P$ preserves equalizers if and only if it preserves coreflexive ones. However, there's a point in the example you chose that I find obscure: $mathbbZ_2otimes ker(a) = mathbbZ_2otimes(0oplus bigoplus mathbbZ) cong bigoplus mathbbZ_2 cong ker(mathbbZ_2otimes a)$.
â Ender Wiggins
Aug 26 at 15:43
What about $M=bigoplus mathbbZ$, $N=mathbbZ_2oplusbigoplusmathbbZ$, $f(x)=([0],x_1,x_2,ldots)$, $g(x)=([x_1],x_1,x_2,ldots)$?In such a case $ker(f-g)=2mathbbZoplusbigoplus mathbbZ$, whence $ker(f-g)otimesmathbbZ_2=(2mathbbZotimes mathbbZ_2)oplusbigoplus mathbbZ_2$ while $ker((f-g)otimesmathbbZ_2)=bigoplus mathbbZ_2$
â Ender Wiggins
Aug 26 at 15:54
What about $M=bigoplus mathbbZ$, $N=mathbbZ_2oplusbigoplusmathbbZ$, $f(x)=([0],x_1,x_2,ldots)$, $g(x)=([x_1],x_1,x_2,ldots)$?In such a case $ker(f-g)=2mathbbZoplusbigoplus mathbbZ$, whence $ker(f-g)otimesmathbbZ_2=(2mathbbZotimes mathbbZ_2)oplusbigoplus mathbbZ_2$ while $ker((f-g)otimesmathbbZ_2)=bigoplus mathbbZ_2$
â Ender Wiggins
Aug 26 at 15:54
1
1
@EnderWiggins The equalizers may be abstractly isomorphic, but saying that the functor preserves them means the natural map is an isomorphism.
â Pedro Tamaroffâ¦
Aug 26 at 16:16
@EnderWiggins The equalizers may be abstractly isomorphic, but saying that the functor preserves them means the natural map is an isomorphism.
â Pedro Tamaroffâ¦
Aug 26 at 16:16
 |Â
show 1 more comment
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
StackExchange.ready(
function ()
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f2894931%2fdoes-the-tensor-product-preserve-coreflexive-equalizers%23new-answer', 'question_page');
);
Post as a guest
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Sign up using Google
Sign up using Facebook
Sign up using Email and Password