The Scientific Organizing Committee is to be commended for expanding this debate to include a wider range of topics that are clearly important for the study of ULIGs. Both sides in the debate would likely agree that ULIGs represent an extremely important stage in galaxy transformations (e.g. through the building of ellipticals from merging spirals) and the accompanying metal enrichment of the intergalactic medium via nuclear superwinds. Both sides would probably also agree that research on local ULIGs has become even more important following the extremely exciting discoveries from ISO and SCUBA that suggest that the far-infrared/submillimeter background radiation (as measured by COBE) may indeed be resolved into a population of ULIGs at z ~ 1-4, and that the far-infrared/submillimeter luminosity from these high-z ULIGs may dominate that attributed to star formation as seen in the rest-frame optical/UV.
It also seems clear that most participants at this workshop believe that both starbursts and AGN are important at varying levels to the luminosity output of ULIGs, and that there is most likely an evolutionary connection between the fueling of starbursts and the fueling of AGN. Indeed, what better time to fuel both a circumnuclear starburst and an AGN than when dumping ~ 1010 M of gas and dust into the central kiloparsec of a merger remnant. And if ULIGs indeed represent the building of massive bulges during the merger of gas-rich spirals, then the recent discovery that all massive ellipticals appear to contain massive black holes, where MMBH is a constant fraction (i.e ~ 0.006) of Mbulge (e.g. Kormendy & Richstone 1995; Magorrian et al. 1998) also suggests that the building/fueling of a massive black hole is likely to be concurrent with the ULIG phase. And even if star formation is still favored as the dominant luminosity source in most ULIGs, members of the starburst camp might consider ``affiliate membership" in the AGN camp simply by subscribing to the hypothesis that most ULIGs may eventually evolve into optically-selected UV-excess QSOs.
The majority of workshop participants on both sides of this debate seemed willing to accept the idea that ``warm'' ULIGs and those objects with the highest infrared luminosities (e.g. log [Lir/L] 12.6) are most likely to be powered by a dominant AGN. Harder to accept was a dominant role for AGN in the lower luminosity, cooler ULIGs like Arp 220, but the fact that many Arp 220-like objects appear to have radio and hard X-ray properties similar to the mean properties of RQQSOs (e.g. Mrk 273, UGC 05101) suggests that dust obscuration could still mask a dominant AGN in the majority of these objects as well.
I am grateful to Jason Surace and Karen Teramura for assistance in preparing the figures, to Aaron Evans, Jackie Fisher, Hagai Netzer, Gene Smith, Sylvain Veilleux, and Min Yun for supplying data used in this summary, to all of the members of the AGN-tower discussion group (the ``green team'') for their comments and suggestions, and to JPL contract no. 961566 for partial financial support.