Brain lesions and multiple realization

William Bechtel and Jennifer Mundale make the following comment regarding lesion studies and multiple realization:

Nevertheless, it is important to note
that in interpreting these deficits, researchers implicitly reject
multiple realization among human brains and assume that damage to a
brain area in anyone will result in a deficit to a particular
cognitive function that is performed by that area in undamaged
brains. (Bechtel & Mundale, 1999, p. 184).

This seems to be an overinterpretation of the implications of
localization research. For simplicity of exposition, let us consider
only one area, say, V1 and suppose, again only for the sake of
simplicity, that it has only one cognitive function. Finally,
suppose that V1 displays some variation in size and position from
individual to individual, but is constant enough to enable
investigators to assume that damage to V1 in one individual will lead
to comparable damage in another individual.
Let us concede what Bechtel and Mundale are pointing out and what
seems to be entirely correct. Let us say that V1 is univocally
localized.
The point to observe now is that this univocal
localization is not the same thing as univocal realization.
Moreover, by itself, buying into univocal localization does not
amount to rejecting multiple realization. For it may still be that
there are any number of ways in which cells of V1 might be organized
to carry out the functions of V1. There might be lots and lots of
different ways of putting together nerve cells to form a region of
brain tissue that performs the function(s) of V1 whatever those
happen to be. Were this to happen, were there to be different ways
of assembling neurons so as to realize the function(s) of V1, this
would look to be a case of multiple realization. So, it appears that one can have
a psychological function be multiply realized, even if it is realized
in only one region of the brain.  So, Bechtel and Mundale do not here have a compelling argument against the multiple realization of psychological functions.


40 Comments

  1. Alex

    There’s an important distinction that you seem to be eliding over here, but which, I think, is crucial to understanding what Bechtel & Munsdale are claiming. That distinction is between multiple realizability and multiple realization. You seem to be addressing the notion of multiple realizability when you make the modal claim that “There might be lots and lots of different ways of putting together nerve cells to form a region of brain tissue that performs the function(s) of V1”. But Bechtel & Munsdale are addressing the issue of multiple *realization* — the question of whether multiple brain state types in *actual biological brains* realize a given psychological state type. This should be made apparent by the following quote, taken from the second paragraph of their paper:

    “Our primary concern… is with the implication drawn from the multiple realizability argument that information about the brain is of little or no relevance to understanding psychological processes. The argument generally relies not on the mere logical possibility of multiple realization, but on the contention that mental or psychological states are multiply realized in existing biological organisms”.

  2. kenneth aizawa

    Alex,
    I don’t think the distinction between multiple realization versus multiple realizability will help Bechtel and Mundale here.  To state my point very briefly, and a bit roughly, it is that univocal localization does not entail univocal realization.  In other words, univocal localization does not preclude multiple realization.  Sorry for the confusion due to the earlier formulation.

  3. Ken,

    You make it sound like all there is to being V1 is being some place in the back of the head. But a lot of what goes into distinguishing V1 from other areas involves considerations like “ways in which cells of V1 might be organized”. Consider, for example, celluar distinctions that give rise to the difference between striate- and extrastriate cortex.

  4. kenneth aizawa

    Pete,
    Of course, I don’t think that all there is to being V1 is its location in the brain.  It seems to be delimited by anatomical and certain physiological properties.   It seems to me that you have picked up on another infelicity in my exposition of my objection to B&M.  Here is another way of putting what I am driving at: Some cognitive function F might be found to be univocally localized in one region of the brain, but that function might still be multiply realized.  That avoids both the realization/realizability problem and the problem of saying just exactly what V1, or any other piece of brain-meat, is.

  5. Alex

    Ken,

    Earlier you seemed to cash out the distinction between univocal localization and univocal realization in terms of the modal notion of multiple realizability. But if we agree that by “multiple realization” we mean multiple realization (as opposed to multiple realizability) I lose my grip on precisely what the distinction between univocal localization and univocal realization amounts to.

  6. kenneth aizawa

    I take it that univocal localization is occuring in (essentially) the same place in the brain.  Univocal realization is something like built in only one way.  Both of these are non-modal.  Or put it this way, a cognitive function is multiply localized in the brain iff it occurs in multiple locations in the brain.  A cognitive function is multiply realized in the brain iff it is built in many different ways in the brain.  Both of these are non-modal.

    I’m thinking that Bechtel and Mundale appear to have a good argument in the passage cited insofar as this distinction is unclear.

  7. Alex

    So we adopt the following terminology: (i) “cognitive function F is univocally localized to brain area A” means that F occurs in a single brain area, namely A (where A is a brain area of some actual biological organism type), and (ii) “cognitive function F is univocally realized in brain area A” means that, of the many ways of implementing F, only one way is actually implemented in the relevant organism type, and it is implemented in A.

    Note the distinction here, reminiscent of Marr’s distinction between computation and algorithm, between a given function F and a particular way of ‘doing’, or implementing, F. We can grant that there might be indefinitely many *possible* ways of implementing a function F, but recall that we’re only interested in the ways that F is implemented in actual brains. Now, if a statement of form (i) is true (for given F, A, and organism type O), then it seems that the only way the corresponding sentence of form (ii) can be false is if F is actually implemented in several different ways, and all of these occur in area A. That seems weird. In particular, it seems beg the question against one of the central claims of the BM paper, which is that brain areas are individuated not only on the basis of their location, but also on the basis of what they do and how they do it.

  8. kenneth aizawa

    Now, if a statement of form (i) is true (for given F, A, and organism
    type O), then it seems that the only way the corresponding sentence of
    form (ii) can be false is if F is actually implemented in several
    different ways, and all of these occur in area A.

    Agreed.  This is what I think can happen.  F is always done by Area A, but Area A uses many different combinations of neurons to get the F job done.  That’s what it is for F to be univocally localized, but multiply realized.

    That seems weird. In particular, it seems beg the question against one
    of the central claims of the BM paper, which is that brain areas are
    individuated not only on the basis of their location, but also on the
    basis of what they do and how they do it.

    B&M do say that in neuroscientific practice brain areas are (essentially) individuated on the basis of what they do (Cf. B&M, 1999, p. 177).  (I think this is too strong.  Sometimes neuroscientists taxonomize this way, sometimes no.)  Grant this.  So, you propose that B&M have an argument that goes something like this:

    B&M: Cognitive function F is univocally realized.
    Ken: How so?
    B&M: F is univocally realized by the part of the brain that does F.

    Or maybe like this:

    B&M: Cognitive function F is univocally realized.
    Ken: How so?
    B&M: F is univocally realized by area blah, blah, blah.
    Ken: What is area blah, blah, blah?
    B&M: Neuroscientific practice shows us that area blah, blah, blah is defined as the area that does F.

  9. Ken,

    Let me try to put my point in a slightly differnt way. I think that your claim that V1 might be an example in which F is univocally localized while multiply realized depends on an assumption that V1 itself is multiply realized. But what I would like to suggest is that V1 is not multiply realized. If this suggestion is correct, then I don’t see how there could be any room for F to be multiply realized while univocally localized in V1.

  10. kenneth aizawa

    Pete,
    What I am proposing is that univocal localization does not entail univocal realization.   It does not follow from the fact that F is univocally localized that F is univocally realized.  To show this, I would have thought that it would be enough to show that even if F is univocally localized, it is still possible for F to be multiply realized.  That is why the modality crept into my initial post.   This matters because it seems to me that B&M are supposing that univocal localization does entail univocal realization.

    My post here is to critique a B&M argument.  It is not to argue for multiple realization of cognitive functions.  I have an argument for the MR of cognitive functions in my Synthese biochem paper.  It is not that I don’t think I have the goods to deliver on this MR of V1; it’s that I don’t think I need to deliver those goods for present purposes.

    How about this?  Being a tailpipe is univocally localized, i.e. a tailpipe is always at the distal end of an automobile exhaust system, but being a tailpipe is multiple realized.  There are all kinds of structures that are tailpipes.    This argues that univocal localization does not entail univocal realization, right?

  11. kenneth aizawa

    Pete.

    Consider my friend Fred Adams, who has red hair.

    I claim “Fred is a mammal” does not entail “Fred has red hair”.

    To establish my claim, I only have to show that it is possible for Fred to be a mammal and not have red hair.  I only have to show that even if Fred is a mammal, it is still possible for Fred not to have red hair.  I don’t have to show that Fred is a mammal and that Fred actually does not have red hair.  I can’t do that, because Fred does have red hair.

    What seems to be making this point more complicated than normal is that we are working in the vicinity of a modal predicate, “multiply realizable,” rather than a non-modal predicate, such as “has red hair.”

    = = = =
    So,

    I claim “F is univocally localized” does not entail “F is univocally realized”.

    To establish my claim, I only have to show that it is possible for F to be univocally localized and not be univocally realized.  I only have to show that even if F is univocally localized, it is still possible for F not to be univocally realized.  I don’t have to show that F is univocally localized and F is actually not univocally realized.

    = = = =
    Doesn’t this seem right?

  12. Buddy

    Kenneth,

    The problem, as I see it, is this: Yes, as a point of logic, it is true that “F is univocally localized” does not entail “F is univocally realized.” It is possible for some function to be univocally localized without being univocally realized. But you could have just as easily pointed out, that, as a matter of logic, its possible that cognitive function researchers make unjustified inferences.

    The point is that the datum in the B&M argument is the fact from scientific practice that cognitive function researchers make inferences from one brain to the next. B&M are simply asking what assumption lies behind this inference and to take that as some indication about what researches in the field assume to be the case about MR. That’s not a point of logic. The fact that its logically possible that these researchers believe in univocal localization without believing in univocal realization does does not force the conclusion that its merely belief in UL that is the best explanation of their practice. B&M seem to think that the best explanation of the practice of making those sorts of inferences is belief, on the part of researchers, in univocal realization. Just pointing out that UL doesnt logically imply UR is not an argument against that claim–even if the claim itself is not necessarily well argued for in the first place.

    One actually has to look at what, in fact, is the background assumption that sanctions these inferences.

  13. kenneth aizawa

    The fact that its logically possible that these researchers believe in
    univocal localization without believing in univocal realization does
    does not force the conclusion that its merely belief in UL that is the
    best explanation of their practice.

    Agreed.  I never said otherwise.  My point has been that this argument, so far at least, does not show UL of cognitive function.

    B&M seem to think that the best explanation of the practice of
    making those sorts of inferences is belief, on the part of researchers,
    in univocal realization.

    Ok.  One way to read what I have been driving at is that B&M need to show more than just that neuroscientists believe in univocal localization.  More needs to be said.  What more is there?  Why is the best explanation of neuroscientific practice here that neuroscientists believe in univocal localization and univocal realization?  Does the B&M, 1999, paper spell out why this is the best explanation?

  14. Buddy

    I’m not sure why they think belief in UR is the best explanation of that practice. [I would have to read the paper-ha!] My only point was that surely they were not claiming that it was the only logically possible explanation of the practice. Your argument only seems to show there is another logically possible explanation, not necessarily a better one.

    The argument for which is the better explanation has to be on the merits, not on logic (with examples like mammals and red hair.)

  15. kenneth aizawa

    Buddy,
    I’m glad you agree with the logical point that “F is UL” does not entail “F is UR”.  With Alex and Pete holding out and no one agreeing with me, I was beginning to worry.

    So, it seems to me that B&M fail to distinguish between “F is UL” and “F is UR”, so that they think that by showing “F is UL” they have established that “F is UR”.  So, I’m proposing that failure to distinguish these two is the root of the problem for this argument.

    Your reconstruction of B&M’s argument, however, is what exactly?  That they know the difference between “F is UL” and “F is UR”, but rely on some tacit premise that warrants the inference from “F is UL” to “F is UR”.  I don’t see anything in their text that suggests a sensitivity to this distinction.  Nor do I detect in their text any appeal to some kind of best explanation argument.  I could be wrong, blinded by my preconceptions.

    Incidently, I was not thinking about your way as a way of saving B&M’s argument.  I was supposing that there might be some sort of strong MR principle that might be refuted just by UL alone.  So, maybe the way to save this argument is to think the MR hypothesis that is refuted is something like.  “F is multiply realized iff F is realized in every combination of anatomically specified brain tissue.”  UL would refute that version of MR.  But, it seems to me that this is not a very plausible statement of MR (which is not to say that no philosopher has every state a version of MR like this).

  16. kenneth aizawa

    Ok.  Well, it would be possible for someone to take the line you propose.  Only, I don’t think B&M did, for reasons stated in my last post to you.  I think they are just not sensitive to the UL-UR distinction, so they think that showing UL establishes UR.

    The mammals and red hair example was just to explain the point of the logic of entailment though.

  17. Buddy

    I have no idea what they were actually thinking, but I think the central question about the cogency of their argument is this:

    Suppose we accept for the sake of argument that the only two possible explanations for the bit of scientific practice they describe are a)that researchers tacitly embrace UL and b) that researchers tacitly embrace UR.

    The question is: Even though UL does not logically entail UR, is it in fact plausible to believe that researchers embrace UL but not UR. You yourself are supposing that two _philosophers_ of mind are not even sensitive to the distinction. How plausible is it, then, that brain scientists are? And how plausible is it that they would be committed to the one without the other?

  18. kenneth aizawa

    Well, I think it has been common ground in this entry that brain researchers tacitly embrace UL.  The question, then, is do they also embrace UR.  So, “the” two explanations for scientific practice have to be that a) brain scientists accept only UL and b) brain scientists accept both UL and UR.

    Now, I’m asking for a reason to suppose that b) is the best explanation for scientific practice.  I don’t think B&M have one.  You aren’t offering one.

    Or maybe you are offering one.  Maybe it is this. 
    1. Brain scientists are definitely committed to UL. 
    2. Brain scientists don’t distinguish UL and UR.
    3. If brain scientists are definitely committed to UL and don’t distinguish UL and UR, then brain scientists are committed to UL and UR.
    ===
    So, brain scientists are committed to UL and UR.

    Is this your argument?

  19. Buddy

    Well, I’m not actually offering an argument, just wondering whether there is one to be offered. One would be the one you give here. Another would be that while UR is logically distinct from UL, UL without UR is an implausible antecedent tacit commitment to attribute to researchers.

  20. Alex

    *Of course* ‘F is UL’ does not entail ‘F is UR’. *Of course* it is logically possible that a given F is UL but that it is not UR. But, to repeat, B&M are not concerned with whether or not this is logically possible; they are concerned with whether or not instances of this are *actual*. That is, they are concerned with whether or not there are any actual cases of a function being localized to a given brain area but nevertheless being realized in different ways within that area. Actually, even this is not quite right. B&M are arguing that, as a matter of fact, neuroscientists adopt the methodological presumption that functions are univocally realized. They are NOT making the much stronger claim that there are in fact NO instances of multiple realization; their point is that such cases, if they exist, are rare enough for the neuroscientist’s methodological presumptions to be useful and well-motivated.

    So to provide a convincing objection to B&M, it is not enough to point out that ‘F is UL’ does not entail ‘F is UR’. Nor would it be enough to show that some function actually is localized in a given brain area but is nevertheless realized in different ways within that area, for B&M do not claim that such cases do not exist. In order to object to their argument, one would need to show that examples of univocal-localization-of-multiply-realized-functions are in fact common enough to make it unwarranted for scientists to make the methodological pressumption of UR.

    As a side note, I find the suggestion that B&M are simply unaware of subtleties concerning the distinction between localization and realization highly implausible, given that Bechtel has done more work on the epistemic role of localization in the life sciences than any other philosopher I can think of. I find it far more plausible that B&M simply don’t mark the distinction in their paper because they consider it entirely tangential to their arguments.

  21. anna-mari

    Hi Ken,

    You have been busy yesterday? Anyway, it was nice to read these posts.

    I have not decided yet, whether or not I agree with you. There is a question I´d like to pose before making any decisions.

    What about this (imaginary) case: It is possible (that a cognitive function C is realized in system S (head in Helsinki) in neuron(population)s that fire between 0-5mV:s and in another system S´(head in, say, Jerusalem) in neurons that fire, say, between 5-10mVs in a “same” anatomically defined region.

    They can be (at least in a certain degree) anatomically realized in the same “place” (i.e. anatomically defined brain region or whatever), but, on the other hand, they are multiply realized.

    But; even if it were accepted that this case would be an example of univocal localization, the problem is that by changing the localization criteria, we will have quite different conclusion. (If we use the functional localization criteria of some sort, and decide that the difference between firing rates is too big, this will not be an example of univocal localization after all.)

  22. Ken,

    Ok, that’s cool. That does seem right. As a point of modal logic I guess I don’t have a beef with it.

    I have a beef only if you are interested in making a claim supported by evidence about brains in the actual world. There’s more to being a brain region than simply being some head-relative spatial location. And there’s good reason to believe that the something more rules out actual world multiple realizations.

    As for an interpretation of what B&M are up to, it’s been too long since I read their stuff to be sure. But my hunch is that (1) they aren’t drawing the bad modal inference you seem to suspect and (2) they would agree with me about the relevance to their claims against multiple realization of low level details in the actual practice of taxonomizing brain regions.

  23. kenneth aizawa

    Alex,
    It seems to me that in this passage:

    researchers implicitly reject multiple realization among
    human brains and assume that damage to a brain area in anyone will result in a
    deficit to a particular cognitive function that is performed by that area in
    undamaged brains. (Bechtel & Mundale, 1999, p. 184),

    B&M are giving something like the following
    argument

    P1 Lots of Fs are UL.

    P2 F is UL entails F is UR.

    C So, lots of Fs are UR

    I agree that P1 is true.  I think we have all agreed to that.  I’m pointing out that P2 is false.  Now, you and Buddy have agreed that P2 is false.  So, we seem to agree that this argument is no good.

    Your reply seems to be that I have misread the argument.  (This seems to be what Buddy is driving at as well.)  Instead, it is something like this:

    P1 Lots of Fs are UL.

    P2 It is a statistical regularity of the actual world that when an F is UL,  that F is UR.

    C So, lots of Fs are UR

    Is this what you propose they are up to?  (This seems to be other than what Buddy was proposing.)

  24. kenneth aizawa

    I figure there must be other arguments that riff on this theme.  I’m not sure any will do much work.  “UL without UR is an implausible antecedent tacit commitment to attribute to researchers.”  Agreed.  But “If UL, the probably UR” also seems to be an implausible tacit commitment to attribute to researchers.

  25. kenneth aizawa

    Anna-Mari,
    I have a stack of papers to grade, so I’ve been procrastinating.

    What you seem to be proposing is that if we fiddle with the functional localization criterion, e.g. by moving away from something like co-ordinates in Talairach space (actually that won’t work, since we are doing comparative work across species), then we can get it to turn out that univocal functional localization does entail univocal realization.  Maybe.  What is the theory of functional localization you have in mind?

  26. kenneth aizawa

    Pete,
    The first thing I’m up to in this post is trying to defeat one argument for UR.  That’s the one I attribute to B&M in my last post to Alex.  I’m also interested in defeating slight variants of this that others have posted.  So, I don’t think there is likely to be a good argument in the neighborhood of this one, but we’ll see.

    I agree that there is more to being a brain region that simply being some head-relatively spatial location.

  27. Tony

    Ken,

    I don’t know any of the neuroscience or the research that B&M refer to. But the passage that Ken quotes

    researchers implicitly reject multiple realization among human brains and assume that damage to a brain area in anyone will result in a deficit to a particular cognitive function that is performed by that area in undamaged brains. (Bechtel & Mundale, 1999, p. 184)

    suggests the following picture to me. If I’m in the business of trying to localize functions (B&R 1993), in the service of understanding how the mechanism works, it’s sensible to assume (or hope), once I’ve got a candidate function in view, that there isn’t significant variation from subject to subject in *how* an area performs a function. That is: I’m suggesting we think of at least some of this research as having two stages: first figure out *where* the function is performed; then, later, figure out how the area does it. The second stage is the one at which you might find significant variation in how the function is performed.

    So you wouldn’t expect someone involved in the localization project, the first stage, to spend a lot of time talking about the possibility of multiple realization. Moreover, I would guess that often researchers don’t yet have the resources (methodologies or tools) to investigate the possibility of multiple realization in the brain.

    So I wonder whether B&M have overplayed their hand here. There isn’t an argument here that researchers implicitly *reject* MR (or UR). It’s rather that they have no commitment about it either way, since it’s nothing they can do anything about.

    (Krebs cycle: suppose it turns out that there are biochemically significant alternative pathways, and suppose it would require a quite different research methodology to figure that out than was originally used to figure out Krebs cycle. That wouldn’t count as an implicit rejection of multiple realizability *or* multiple realization. It would simply be silence on the issue. )

    –Tony

  28. anna-mari

    Hi,

    What I actually meant was that (i) this is the only case of univocally localized, but multiply realized cases I can imagine. But then (ii) we have to decide, whether or not the case I am offering to you is actually about univocally localization. From anatomical point, yes, but from the functional – well, it depends…

    I was thinking only the usual “oscillatory frequence” (is this the right term in English?) theory of functional localization.

    I should also add that the example I was thinking is inspired by an example proposed by Shagrir in his 2001 paper. He is not talking about localization, but uses his example for other purposes. I just thought it might fit here too… Anyway, a great paper, and a very wise man from east (Shagrir, I mean).

  29. kenneth aizawa

    Thanks, Tony.  This seems to very nicely complement what I am thinking.  With all the resistance, I’ve been wondering whether I’m completely out to lunch.

  30. kenneth aizawa

    Well, here is a case of what seems to me to be UL and MR of a biological structure.  The outer segments of rods in the human retina.  There is a great deal of uniformity in their localization, but lots of different ways of making a rod outer segment.  There are dozens of known different forms of rhodopsin.

    I think that your notion of functional localization in “oscillatory frequency” is probably not what B&M are up to.

  31. Buddy

    I dont see how you could think either one of those is a reconstruction of B&M’s argument. They are not beginning with anything like what you call P1 in both reconstructions.

    Their first premise (p1) is that congnitive deficit researchers make certain kinds of inferences that have a hidden premise. (Hence the words “implicitly” and “assume”–what role do those words play in your reconstruction?)

    It could be that c1 is: the hidden premise is UR.

    to that you could reply that c1 is unwarranted. only c1* follows: the hidden premise is UL with or without UR.

    then B&L would need p2: its implausible that brain researchers would hold UL as a hidden premise without holding UR.

    and then c1 would follow.

    What I was suggesting is that I can think of many arguments for p2.

    1-brain researchers probably dont distinguish UR from UL

    2-UR is the best explanation of UL

    3-Belief in UR is the best explanation of belief in UL

    4-the fact that F is UL is so often associated with F being UR, that it is implausible that they believe in UL without UR, etc.

    In any case: I dont see how you can think premise 1 of the argument is: Lots of Fs are UL, when they say nothing of the kind. The premise is clearly about implicit assumptions required to explain a scientific practice among brain researchers.

  32. kenneth aizawa

    Ok.  Let me step back.

    1.
     Do we agree that B&M want to argue
    for the conclusion:

    C:
    Researchers implicitly reject multiple realization among human brains.

    2. Do we agree that B&M use as one premise for this
    conclusion:

    P1: Researchers assume that damage to a brain area in anyone
    will result in a deficit to a particular cognitive function that is performed
    by that area in undamaged brains.

  33. Hi. In the future I’m going to keep here links to their sites. But I do not worry about the sites where my link is removed. So if you do not want to see a mountain of links, simply delete this message. After 2 weeks, I will come back and check.

  34. How are you. Ask not what you can do for your country. Ask what’s for lunch.
    I am from Sweden and learning to write in English, give true I wrote the following sentence: “When your body fails to produce enough of those.”

    Thanks for the help ;-), Robbin.

  35. Sorry. If you watch a game, it’s fun. If you play at it, it’s recreation. If you work at it, it’s golf.
    I am from Barbados and learning to read in English, tell me right I wrote the following sentence: “Agency name, acts — alcohol chemical treatment series.Apr a website dedicated to help people recover from drug and alcohol addiction.”

    Thanks :(. Yehudi.

  36. Hi guys. All that is human must retrograde if it does not advance. Help me! It has to find sites on the: Order synthroid, requip no rx required, buy tulasi, cheap lynoral without perceived shortage, its use of order synthroid lung, liver kidney.. I found only this – turbo tax. Synthroid. Be associated with hair loss if your thyroid disease is not being well controlled. Below are synthroid levothyroxine reviews, ratings, comments submitted by patients and caregivers. Best regards :eek:, August from Armenia.

  37. Hi everyone. Treat all disasters as if they were trivialities but never treat a triviality as if it were a disaster. Help me! Can not find sites on the: Turbo Tax. I found only this – turbo tax. When your body fails to produce enough of. Compare levothroid versus synthroid. Thank you very much ;-). Lark from Samoa.

  38. Hey. We make a living by what we get, we make a life by what we give. Help me! Please help find sites for: Modern baby nursery furniture, modern baby toys and modern high chairs.. I found only this – turbo tax. For example, many people enjoy baby bedding that is littered with designs of the. Home about faq checkout baby bedding set nature friends. With respect ;-), Bluma from Republic.

Comments are closed.