The lifeworld is a "commonly shared world"; it is "bounded by the totality of interpretations presupposed by the members [of a community of speaking and acting subjects] as background knowledge" (TCA1 p. 13). On the basis of the lifeworld as expressed in these terms, the members of a community have grounds for examining discrepancies in the accounts rendered by two of its members and determining that one account is accurate while the other is impaired.
For example, if two people who appear to be looking at the same object claim to see something radically different from one another, investigation may reveal that one of them is experiencing a problem in eyesight, cognition, perception, accuracy of measurement, bias, or inattentiveness. It seems, however, that limits of time, interest, awareness, and other resources render it impossible, in principle, for a community even to identify, much less to sort out, all instances of discrepant impression held by its members. Moreover, as I understand the Duhem-Quine thesis (and as experience suggests), there may often be no standard by which discrepancies can be identified and resolved; the proposition-based sense of resolution neglects the possibility that beliefs tend to be part of a web that is capable of indefinite amounts of adjustment rather than any simple, Darwinian survival of the strongest idea or view. That sort of conflict of ideas might seem to augur the division into two communities; but the number of such potential divisions is such as to make each man an island rather than the community that Habermas envisions.
Communities could be construed as extending only to those who share a particular view; but within each view are more views -- yielding, again, an archipelago rather than a continent. In short, the concept of a commonly shared community seems to arise, not from an actual sorting-out of differences, but from the assumption that the members could theoretically sort out some such differences, if forced to do so for some reason.
The lifeworld is said to be non-falsifiable because it is not "a descriptive assertion" (TCA1 p. 13). In seeming agreement with the atomistic prospect anticipated in the preceding paragraph (above), it is "the unthematically given horizon within which participants in communication move in common when they refer thematically to some [single] thing in the world" (TCA1 p. 82). As a given, it is not possible for someone living within it to step outside it and render objective judgment on it. Doing so would seemingly amount to an affirmation that the observer is not within it -- that it is not, for him/her, a given.
Habermas seems to think that it is possible, however, to adopt an internal position with respect to another culture, so as to understand it from the perspective of those who are within it (Carspecken, p. 7). Unlike Husserl’s transcendental phenomenology, however, Habermas’s reconstructive analysis employs not only reflection but also empirical analysis. This seems to be a quintessentially modernist perspective. Certainly one can compare cultures -- can even become immersed in them, can "go native." But that one researcher could be immersed in two separate cultures so completely as to have bracketed out the preconceptions from his/her own membership in a third one seems unlikely.