The designation of an ortho-, meta-, or para-substitution pattern was coined when spectroscopic analysis was just starting (Robert Bunsen, for example). Some of the spectroscopic techniques to assign substitution patterns (IR-, and more importantly, NMR spectroscopy) were simply not available. To quote wikipedia's entry about origines of arene substitution pattern:
The prefixes ortho, meta, and para are all derived from Greek, meaning
correct, following, and beside, respectively. The relationship to the
current meaning is perhaps not obvious. The ortho description was
historically used to designate the original compound, and an isomer was
often called the meta compound. For instance, the trivial names
orthophosphoric acid and trimetaphosphoric acid have nothing to do with
aromatics at all. Likewise, the description para was reserved for just
closely related compounds. Thus Berzelius originally called the racemic
form of aspartic acid paraaspartic acid (another obsolete term: racemic
acid) in 1830. The use of the prefixes ortho, meta and para to distinguish
isomers of di-substituted aromatic rings starts with Wilhelm Körner in
1867, although he applied ortho prefix to a 1,4 isomer and the meta prefix
to a 1,2-isomer. It was the German chemist Karl Gräbe who, in 1869,
first used the prefixes ortho-, meta-, para- to denote specific relative
locations of the substituents on a di-substituted aromatic ring (viz,
naphthalene). In 1870, the German chemist Viktor Meyer first applied
Gräbe's nomenclature to benzene. The current nomenclature was
introduced by the Chemical Society in 1879.
Besides many among the chemists are familiar of this designation, it is limited to describe the relationship of two substituents of a benzene. Hence frequently chemical suppliers and data bases prefer to describe the position of them by numbers. IUPAC's Blue Book still allows the use of o, m, and p (rule A-12.3), yet the consent moves into the direction of a consistent numbering that is more adaptable in terms of number of (potentially different) substituents and their relative arrangement to (hetero)aromatic nuclei.
To illustrate this, think of examples like lutidine (systematically 2,6-dimethylpyridine), a frequently used mild, sterically hindered base that does not fit into the other system.
In the end, we've to learn understand both dialects + some of non-systematic names, too.
