A survey on gender systems in 256 different languages shows that 112 (44%) have grammatical gender and the other 144 (56%) are genderless. In most cases, these two types of languages are geographically close together, indicating that gendered languages are attributed to language families at large rather than sporadically attributed to singular languages.
Grammatical genders go further than simply ‘male’ or ‘female’. Finno-Ugric languages distinguish between animate and inanimate objects. Efforts to reconstruct Pro-Indo-European language, a reconstruction of Indo-European languages, the most widely spoken language family in the world, indicate that the animate/inanimate distinction preceded male/female. This clearly fits in with a time where the natural world formed our core beliefs. As monotheistic faith replaced the animistic beliefs of the surrounding world, grammar went through a similar transition.
While the distinction between animate and inanimate was an important one to make in previous societies, as monotheistic faith grew increasingly popular in society, it was even more important to distinguish between feminine and masculine roles.
20th century linguist Antoine Meillet supports this idea stating that gender wasn’t useful in previous times. To back this up, he uses the variety of genders used for the same words in different languages. He says that this proves there is no one characteristic that assigns an object to a specific gender. For example, why should a key be feminine in French (la clé) by masculine in German (der Schlüssel)?
Linguists Arent J. Wensinck and Jean Marakle are some scholars to present this idea, but whether it is simply a coincidence is a strong counter-argument.