How are language families different from language stocks in South American linguistics?

In South American linguistics, the distinction between a language family and a language stock is currently difficult to define with precision due to limited linguistic data and the lack of standardized classification criteria.1

Generally, these terms are categorized as follows:

• Language Family: This is composed of a group of individual languages that share a common origin.1
• Language Stock: This is a broader, more inclusive grouping that consists of multiple language families or, in some cases, highly differentiated languages that share a more remote genetic relationship.1

Because researchers often rely on the inspection of vocabularies and structural similarities rather than the rigorous application of the comparative method, the boundaries between these levels remain approximate. Many recognized groupings in South America are internally diverse, and in some instances, the lowest components identified in classifications are already at the level of stocks or even more distantly related groups. Consequently, a true taxonomy is not yet feasible, and many groups are categorized at various, undetermined levels of relationship.1