Past Mapping Periodic Table

This discussion is about the different ways in which we map the past.

As we all know, every map is a conceptual conceit in which certain aspects of reality are distorted in order to better represent other ones. This is particularly the case in past mapping where often the data that we are dealing with is extremely limited whilst we are (generally) forced by the 'total' nature of spatial representation to give a 'complete' picture. The now common use of multiple layers within a GIS which may have been derived from entirely different contexts only adds to the danger that assumptions may be made, and conclusions drawn that cannot, and were never intended to be, supported by the data.

It is hoped that by listing, and then comparing, the different kinds of information we map, it is possible to build up a kind of family or 'periodic table' of techniqiues which we can then use to appraise the manner in which we undertake these tasks, as well as the ways in which we can meaningfully compare and analyze them. In all cases the point of criteria by which we can distinguish different types of data is "what would it mean for this map to be correct?". i.e. on the assumption that all maps distort elements which are not the primary focus of the map, what must remain undistorted in order to maintain a 'correct' past map?

To use the technical jargon of sufficiency and necessity, the map is accurate if and only if (iff) x. The headings below are intended to generate discussion are not (yet) exhaustive or mutually exclusive. Some of the headings will almost certainly correspond to 'clusters' within the table.

Present Maps

In one sense, present mapping is past mapping also, for there is always an intervening period between an observation and its record. However, present maps are accurate iff a survey carried out now would correspond to the depiction on the map where 'depiction' is shorthand for the representative framework used by the cartographer. In other words, Present Maps can suffer ontological obsolesence (i.e. due to changes in the world). Past Maps can only suffer from epistemic obsolesence (i.e. due to changes in our knowledge).

Past Maps

• Old Present Maps

As mentioned above, all Present Maps suffer from virtually instantaneous obsolesence unless continuously maintained. That makes them useful as 'snapshots' of the past, as long as they can be dated accurately. The map is correct iff the world corresponded to the depiction on the map at the given time.

• Location maps (known points)

• Location maps (known lines)

• Location maps (known polygons)

• Hard boundaries

• Soft boundaries

• 'Natural' variables

• 'Modern' concepts (eg Caucasians?)

• 'Ancient concepts (eg Barbarians?)

• Predictive Modelling

• Viewshed Analysis

• Network Analysis

• Choropleth mapping

• Topological maps/data

• Hodological space

• Universal accuracy (accuracy to an overall framework)

• Local accuracy ( accuracy with regard to eg nearest neighbour)

• Framework accuracy (correspondence of framework to real world)

• Descriptive space(?) (eg Pliny/Strabo etc)