A period dated from indirect clues
The Vedic period has no inscriptions, king lists or datable monuments of its own, and the Vedic texts do not give calendar dates. Historians therefore rely on a combination of clues, mainly language, outside references and archaeology, to place it in a sequence. The results are best treated as approximate. A common outline puts the Early Vedic period, associated with the Rigveda, in the later second millennium BCE, and the Later Vedic period in roughly the first half of the first millennium BCE, though specialists disagree on the edges.
Clue one: the language of the texts
Vedic Sanskrit changed over time, and the texts can be ordered by their grammar and vocabulary. The Rigveda uses the oldest forms. Later collections, such as the Yajurveda, Atharvaveda and the Brahmanas, show newer features, and the early Upanishads are later still. This produces a relative order, which says which text is older but not by how many years. Sanskrit grammar was later fixed by the grammarian Panini, traditionally placed around the fourth century BCE, though his date is also debated, and this gives a rough lower limit for Vedic usage.
Clue two: outside references
A few external records help. A treaty between the Hittite king and the ruler of the Mitanni kingdom, usually dated to the fourteenth century BCE, invokes gods whose names resemble Vedic Mitra, Varuna, Indra and the Nasatyas. Names and horse-training terms in Mitanni sources also have Indo-Aryan features. These show that people with an Indo-Aryan linguistic heritage were active in the Near East by that time, although they do not say where or when the Rigveda itself was composed. The historical Buddha, whose dates are also debated but are often placed in the fifth century BCE or thereabouts, lived in a world that already knew the later Vedic literature.
Clue three: archaeology
Archaeologists try to match texts to material cultures, but matches are uncertain because texts and pottery do not label each other. Painted Grey Ware, a fine grey pottery with black designs found in the upper Ganges plain and Haryana, is often dated to about 1200 to 600 BCE and is commonly linked with Later Vedic communities. The distribution of iron tools, which appear in northern India in the early first millennium BCE, fits the iron mentioned in later texts. Sequences of pottery styles, studied through pottery seriation and calibrated radiocarbon dating, give the main framework. Linking a text to a particular pottery culture, however, is an interpretation, and it is argued over.
Early and Later Vedic phases
The earlier phase is represented by the Rigveda, composed orally and collected over some centuries. Its setting is the northwest and Punjab, with seven rivers, pastoral clans, cattle, horses and chariots, and a ritual religion centred on fire and the soma drink. The later phase, shown in the other Vedas and Brahmanas, shifts east to the Ganges-Yamuna region, with more settled agriculture, larger political units and more elaborate ritual specialisation. The early states and cities of the middle Ganges, which come at the end of this sequence, are generally dated to the middle of the first millennium BCE.
The unsettled link to the Harappans
A long-standing question is how the Vedic communities relate to the earlier Harappan civilization. The Rigveda does not clearly mention Indus-style cities, and the Indus cities were already in decline by about 1900 BCE, several centuries before the usual dates for the Rigveda. That leaves a gap, and different scholars fill it differently.
The mainstream linguistic view holds that Indo-Aryan languages spread into South Asia from the northwest with groups arriving in the second millennium BCE, perhaps in a series of movements over time rather than a single event. Older versions of this idea, which pictured a violent invasion that destroyed the Indus cities, are no longer supported; the archaeological evidence for such destruction is lacking. Evidence from ancient DNA has been used in support of migration, since some studies report that steppe-related ancestry appeared in South Asia in the second millennium BCE. A 2019 study of a skeleton from Rakhigarhi, a large Harappan site, found no such ancestry, which has been widely discussed.
Other scholars argue for greater continuity between the Indus and later Vedic cultures, pointing to features such as ritual, river names including the Sarasvati, and the lack of a single obvious break in the archaeological record. Critics of that view note that horses and spoked-wheel chariots, central in the Rigveda, are weakly attested in Harappan sites. Genetic results cannot by themselves show which language people spoke, and archaeological cultures do not map neatly on to languages or peoples.
How to read the debate
It helps to separate three questions: when the texts were composed, where and how Indo-Aryan languages arrived, and whether the Harappans spoke an ancestor of those languages. They can be answered separately, and the evidence for each differs. At present, a reasonable summary is that the Vedic period is dated in broad terms to roughly 1500 to 500 BCE, that the Harappan connection is unresolved, and that claims presenting any one theory as proven go further than the evidence does.