In 1995, D. Lavallée published Promesse d’Amérique. La Préhistoire de l’Amérique du Sud, a monograph that is still a reference today. She concluded that first human occupations in South America would have arrive as a slow and irregular migration that could have started at least thirty thousand years ago. D. Lavallée offered in this work one of the rare objective visions of the South American archaeological record, considering in a neutral way the various hypotheses present in the 1990s. More than twenty-five years later, no other French-language publication has attempted to replicate the same ambitious project about the prehistory of South America. Far from this ambition, this work presents a new global vision of the first human occupation of this subcontinent, in the light of new archaeological, paleoenvironmental and genetic data which enrich our understanding of the technical and cultural phenomena produced between the late Pleistocene and the early Holocene. This updated vision makes it possible to define a relatively deep prehistory for South America which begins at least 50000 BP, long before the Last Glacial Maximum (LGM; ~26500-19000 cal. BP). Pleistocene records in South America are increasingly enriched with new data. Seven sites are dated to the pre-LGM (55000-29000 BP), 5 sites to the LGM, and more than 30 sites to the post-LGM (19000-12800 cal. BP). Towards the beginning of Marine Isotopique Stage (MIS) 3, between 60000 and 50000 BP, four sites are recorded in the eastern intertropical region of South America, northeastern Brazil and the Argentine pampas. These sites present virtually the same chronology and similar archaeological records (simple debitage systems associated to unifacial and bifacial shaping on pebbles and cobbles) occupying different biomes at a time when the North American ice sheets have not yet merged and the Beringia land bridge is intermittently available, as well as the pacific coastline free of glacial cover. At the end of MIS 3, about 30000 BP, when the North American ice sheets began to merge, we count about ten archaeological sites virtually in the same biomes as at the beginning of this isotopic stage, sharing the same previous technological trends but with the addition of new tools, in particular convergent tools on pebbles and cobbles. At the beginning of MIS 2, about 29000 BP, and especially between 26000 and 24000 BP, when the Beringia land bridge was fully available, but the North American ice sheets are fully merged, we observe the presence of the Patagonian ice sheet, in the extreme south of South America. During this period, the spatial pattern of archaeological sites changed in the intertropical zone of South America, from north-south (or vice versa) in the pre-LGM periods to an east-west direction. A microlitization phenomenon is observed at some sites during this period. At the end of the LGM and immediately after, no major changes are observed in the number of sites, but new biomes are occupied especially on South America’s borders. The bifacial phenomenon and especially the projectile points begin to emerge during this period. Immediately after, 15000 BP, within continental deglaciation, the number of sites doubles, following the same spatial distribution as the previous period. The bifacial phenomenon is increasingly present, but apparently not in the intertropical zone of South America, where the so-called “choppers” and simple flakes industries continue to persist. About 13,000 years ago, the North American ice-free corridor was available and a demographic explosion is observed across the continent. From this period, virtually all South American biomes are occupied by humans. We also observe that projectiles points acquire an archaeological visibility without precedent. This great diversity of South American archaeological records finds a logical explanation only if we hypothesize a Pleistocene occupation modulated by repeated cycles of Eurasian demographic expansion-diffusion from the end of the MIS 4, particularly in four palaeoenvironmentally optimal entry intervals: 55000-60000 BP; 37000-40000 BP; 26000-30000 BP, and 16000-11000 BP. Consequently, multiple demographic expansions would be responsible for the main South American technical phenomena recorded to date. Generally, across South America, the unipolar flake debitage, bipolar on anvil debitage and unifacial shaping precede the emergence of bifacial shaping associated with the fabrication of projectile points, a clearly post-glacial phenomenon. The post-Bølling-Allerød/Antarctic Cold Reversal period marks the transition from the Late Pleistocene to the Early Holocene, a period in which a wide variety of regional technical phenomena emerge in almost all parts of South America at an unknown rate in evolutionary histories from other parts of the world. Archaeological and paleoenvironmental data in South America thus reinforces a model of long chronology proposed many years ago but rarely fully developed. In the light of these observations, the general outlines of D. Lavallée’s proposal (1995) for the prehistory of South America do not seem to undergo major changes, except for a new chrono-cultural framework. Twenty-five years after the publication of her monograph, South America pre-LGM peopling is no longer a promise, but a reality.