Why did some scholars once propose an especially close relationship between Celtic and Italic languages?

Some scholars once proposed an especially close relationship between Celtic and Italic languages, even suggesting the existence of an "Italo-Celtic" nation toward the end of the 2nd millennium BCE. This hypothesis was primarily based on shared linguistic features, such as:1

• The q-p relationship: Both branches exhibit variations in how they handle the Indo-European labiovelar stop kʷ, with some Celtic and Italic languages shifting the sound to p while others retained a q or c sound.12
• Morphological similarities: Scholars pointed to specific verb formations, such as the deponent and passive endings in -r, and certain superlative endings for adjectives (e.g., Latin -issimus and Celtic -samos).1

However, this theory has been heavily criticized by linguists like Carl J.S. Marstrander and Calvert Watkins. Many of the cited similarities are now considered accidental resemblances or independent developments. For instance, the -r endings have since been identified in other Indo-European branches, such as Hittite and Tocharian. Consequently, modern scholarship generally views these commonalities as the result of convergence or independent processes rather than evidence of a unique, shared ancestral branch.12