Sandel's two examples are interesting both in themselves and in terms of the way they inflect his general question about patriotism. About a decade ago Pauline Kleingeld published an article in Philosophy and Public Affairs which responded to the general challenge Sandel has put again in his recent interview and did so by invoking the notion of "Kantian patriotism".
Since Philosophy and Public Affairs operates a pay-wall it is probably worth giving a quick summary of Kleingeld's argument in order to test its use as a response to Sandel. In brief, Kleingeld points out that patriotism does not have only one sense. She distinguishes between three different senses of the term. The first is "civic patriotism" which has an intrinsically political nature being concerned with shared political freedom and the institutions that sustain it. Such a form of patriotism is not based on ethnic traits but can embrace a reference to national traditions if they are based on political ones. This view is contrasted by Kleingeld to "nationalist patriotism" which has a key reference to a national group incorporated in it although such a group does not have to be one of shared ancestry and, importantly, the boundaries of the nation need not coincide with those of a state. However, embodied in this second view is an important comparison between the nation and the family so the lack of choice about belonging to a nation is, somewhat paradoxically, part of what one values in it. Finally, there is trait-based patriotism, which is inspired by love of certain qualities. This latter view can inspire love of a people not one's own though in such cases it is not usually called patriotism.
After distinguishing these three types of patriotism Kleingeld goes on to argue that the structure of Kant's arguments in the Doctrine of Right and in Perpetual Peace for republicanism should lead Kantians to embrace an identity as active citizens who are concerned to uphold and enhance traditions of engagement with the polity. This effectively leads her to endorse civic patriotism due to its lack of exclusivity.
There are some useful lessons from Kleingeld's analysis. It is clear, from thinking through the differences between these three different kinds of patriotism, that it is often the case that a given commitment to patriotism often partakes of mixtures of them. So Sandel's reference to American patriotism and its embrace of the constitution and a political history, is not purely a civic patriotism but also embodies trait-based patriotism and can, for some at least, be also a form of nationalist patriotism. The German case is more difficult because it is here a special set of obligations that are said to lie on the people in question due to crimes of their ancestors. This embodies a sort of nationalist patriotism but not one of pride in one's ancestors but a kind of shame in them. It inverts the nationalist case and tries to use it to instil a civic patriotism.
The German case involves a kind of addition of an "incentive" in Kant's sense on top of the normative commitment to republicanism. There are obvious reasons why this case arose but it also entails that the relationship to the constitution in question is not purely one of a civic sort.
Other examples could also be used to complicate the discussion. The example of South Africa would be one given the "truth and reconciliation" notion as a way of attempting to deal with a political history of conflict in which the majority were subordinated. It is not obvious, in some respects, however, whether this project is really as viable as it first seems.
In any case, what does emerge from combining Kleingeld's analysis with Sandel's is that the Kantian commitment to republicanism does contain special duties of citizenship in it. These special duties are sufficient to mandate a patriotism of a sort but the relationship to the other forms of patriotism is more difficult. Kleingeld effectively rules out the view that a Kantian could endorse either nationalist or trait-based patriotism but she does not sufficiently discuss the question of how these latter two forms of patriotism incorporate within them reasons for supporting traditions that could be civic as well as reasons for supporting non-civic identities.
The basic problem with nationalist patriotism, it seems to me, of the sort implied by Sandel's reference to America, is that it can involve a selective sense of the moral history of the nation in question. Whilst it is true that there are morally inspiring elements of American history there are many elements of that history that are rather darker including the founding genocide of the indigenous inhabitants, slavery, and international conduct that has often involved domination of others for a purely national interest. This is not to indulge in anti-Americanism since much can be said that is far from complimentary about most states. It is, however, to indicate a problem with national stories since they invariably only present nations in a good light, leaving aside that which is questionable.
Similarly, trait-based patriotism is open to the same objection. This does not, however, forestall entirely the point of Sandel's comment. The question of relations that are particular in their import is a salient one. It is just that the basis of these relations being the ones taken particularly seriously should not primarily rest after all on the cases of nationalist or trait-based patriotism.
The German case suggests a problem with dismissal of considerations that arise from nationalist matters (even if they are here in an inverse form to usual). But what this case shows is that the commitment to the civic becomes part of the identity of a people and that, after all, is compatible with a general civic view of patriotism. Indeed, it indicates reasons for generalizing it.