A macrofamily (also called a superfamily or superphylum) is a term often used in historical linguistics to refer to a hypothetical higher order grouping of languages.
Metonymically, the term became associated with the practice of trying to group together various languages and language families (including isolates) in a larger scale classification.[1][2] However, some scholars [3] view this term as superfluous if not outright redundant as there is no real tangible linguistic divide the same way there is between a linguistic isolate and a language family proper.
Lyle Campbell, professor at the University of Hawaiʻi at Mānoa, had famously said that he is preferring to use the terms "language family" for those classifications for which there is consensus and "distant genetic relationship" for those for which there is no, or not yet, consensus, whether due to lack of documentation or scholarship of the constituent languages, or to an estimated time depth thought by many linguists to be too great for reconstruction.[citation needed]
Examples of proposed macro-families[1][4] range from relatively recent such as East Asian, Macro-Jê, Macro-Waikurúan, Macro-Mayan, Macro-Siouan, Penutian, Dené–Yeniseian and Congo-Saharan (Niger-Saharan) to older ones such as Austric, Dené–Caucasian, Eurasiatic, Nostratic, Borean or Ural-Altaic.
Sometimes the term has also been applied to very old, large and diverse language families, such as Afro-Asiatic and Sino-Tibetan.[5]