I always assumed that this terminology originates from computer science, as in how relational databases organize data into records (rows) that consist of many "fields" describing attributes. I haven't even considered other possibilities you mention.

I just checked the reference manual of CGIS from 1967 [1] where they refer to it as "descriptor" (see e.g. page 17, Data Bank Organization). It is kind of expected since the relational database model was first proposed in the early 70s, I believe.

My guess would be that "field" to describe geographic attributes was first used by one of the early computer cartographers in the 70s and they adapted the term from CS. The journal "Cartography and Geographic Information Science" dates back to 1974 when it was called "The American Cartographer". It would be interesting to look back at its first volumes and see when exactly this terminology became widespread (or more precisely how long it took to for "field" to replace other terms like "attribute", "variable", "descriptor", etc.)

[1] R.F. Tomlinson 1967. An introduction to the geo-information system of the Canada Land Inventory - https://gisandscience.files.wordpress.com/2012/08/3-an-introduction-to-the-geo-information-system-of-the-canada-land-inventory.pdf

Thank you, Levente! Yes, "field" is used to refer to the components of database records (as well as scientific fields, agricultural fields, field work etc etc). However, I am talking about the (originally) physics notion of a field, as in electromagnetic field, gravity field, wind field etc.

My best guess for the earliest treatment *in geography*, so far, is in Peter Burroughs' book (first edition, 1986) https://s3.amazonaws.com/academia.edu.documents/2438559/9fjg8q78n4wux4l.pdf?response-content-disposition=inline%3B%20filename%3DPrinciples_of_geographical_information_s.pdf&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIAIWOWYYGZ2Y53UL3A%2F20190626%2Fus-east-1%2Fs3%2Faws4_request&X-Amz-Date=20190626T144039Z&X-Amz-Expires=3600&X-Amz-SignedHeaders=host&X-Amz-Signature=e27c31d15ebc1c2fe50c6779d9d1db70ebdb9f7613c98f2fcada0f3e18e2f316

I still need to trace his references though.

Not a direct answer, but in [1], p. 243:"It has been traditional in geography to speak of point-to-area (and area-to-point) movements in terms of fields such as a city's sphere of influence..." and "When a point interacts with an area, a field is established. Some fields are theoretically continuous, with interaction rates declining regularly with distance from the center."

A very interesting book anyway.

Another, still not that formal treatment is in [2], one of the earliest I can access, and the terminology is still not very settled (conflates this with rasters a lot). I cannot access the Edition 1 of Burrough, the 98 editioon refers back to Couclelis for this, but that is clearly not the earliest use.

[1] Abler, R., Adams, J. S., & Gould, P. (1971). Spatial organization: the geographer's view of the world.

[2] GOODCHILD, M.F., 1989, Modeling error in objects and fields. In Accuracy of Spatial Databases, M.F. Goodchild and S. Gopal (Eds), pp. 107–114 (London: Taylor & Francis).

Good luck, very interested (specially knowing why you are asking :)

Also, Peuquet 10.3138/D794-N214-221R-23R5 does not directly use the term field, but talks about data models and talks of spatial tesselations. The distinction must have been made somewhere around this time, though.

Werner, this also seems of interest:

S. Angel, G. M. Hyman (1976) Urban fields : a geometry of movement for regional science, Pion.

https://trove.nla.gov.au/work/11047325

Mostly applied to what they call velocity fields, and do not relax the requirement - assume continuous functions, and do not discuss nominal attributes. Mathematically rigorous, though.

'Fields' is a pretentious word used to replace the prior 18th century expression 'action at a distance' which is more honest since it does not pretend to know the answer. Magic spells would also be more honest.

When we see magnetism flowing with the current we start seeing reasons for events. Max can measure magnetism but his vector is absurd.