There is no definite answer. The first scientist that did write on the origin of the Albanians was Johann Erich Thunmann. In his Research on history of East European Nations (Leipzig, 1774) he wrote: In their /Albanian/ history I didn't come across on traces of any latter migration; their language carries such testimony on destiny of the people, that is was impossible not to recognize ancient neighbours of the Greeks and slaves of the Ancient Rome. Both of them guide us towards Ancient Illyrians. This thesis was accepted by Johann Georg von Hahn, who repeated this in his Albanian Studies (Jena, 1854). Due to this work, Von Hahn is deemed the father of Albanian studies. Von Hahn considered the Albanians to be descendants of the ancient inhabitants of the Balkan peninsula (same as the Pelasgians). He tried to demonstrate his claim with help of geographic nomenclature. Such attempts are outdated, but his thesis on Albanian autochthonous origins remained.

Georg Stadtmüller proposed the now-dominant theory in his Research on Early Albanian History (Budapest, 1942). Stadtmüller claimed that the Albanians had developed from certain ancient Balkanic relict that remained non-Romanized. He took the starting point of the formation of the Albanians in 200 BC, and claimed that nothing about that period was known (concerning the /pre/-Albanians, that is). He split the period of the formation of the Albanians in two phases: proto-Albanian (200 BC - 600 AD) and (600 AD - 1050 AD). The proto-Albanian period was marked by expansion of Roman culture and gradual extinction of various Balkanic cultures. He thought that the ancient Balkanic cultures managed to survive only in mountainous areas that were hardly accessible. Therefore, Stadtmüller concluded that, based on geography, the cradle of the Albanians was around the river Mat. The process of Roman domination was suddenly ended by Slavic migration in the 6th century - and in such way the Slavs saved the Albanian culture from Romanization, in a way. Of course, a degree of Romanization is iminent: therefore in Albanian city is qytet, prince is mbret. Stadtmüller supported his theory on the fact that the region around the Mat had the least amount of Slavic toponyms in comparison with other parts of Albania. As you might have noticed, I said nothing about the Illyrians. That's because Stadtmüller evaded this question, saying that the ancestors of the Albanians were possibly one of Illyrian or Thracian tribes.

In newer times Gottfried Schram put a new (or rather, revived) claim that the Albanians originate from the Thracians (more precisely, the Bessi) in his work The Beginnings of Albanian Christianity. The Early Conversion of the Bessi and its Consequences, Freiburg 1994). However, there are no strong arguments in this favor as language of the Bessi is practically unknown (same goes for Illyrian as well, though).

As far as the Albanians themselves are concerned, their common narrative gives them no place for doubt that they are autochthonous, not only in one region of what's today Albanian ethnic area, but on the entirety of it. Neither the Romans nor the Slavs could dissolve the continuity of the Illyrians/Albanians and they had no authority outside bigger cities (some Albanians deny even this). Of course, this image is simplified, uncritical and constructed.