Ever since wizard schools appeared in RPGs, it has been understood that training starts in childhood because teaching and learning so much takes time. Someone immediately remarks that children could learn reading and writing elsewhere, and are too young for the skills more important to spellcasting. Then come characters who, for one reason or another, supposedly need no ancient language or rune magic; instead... The other day I began running a campaign following the characters' lives. Not from first level, but from their first days at school. It is a boarding school, and everyone is still crying for their parents. Their shared cultural background amounts to broken Common.

On the principle of tolerating other religions' followers if they tolerate you, I doubt anyone wanting to send a child to the school would be excluded over cultural or religious differences. Three important subjects immediately appear: learning to work with others; expanding vocabulary by learning broken Common from one another; and the basics of the mosaic principle, one of wizard training's most complicated parts. Teaching future wizards this aspect of their outlook takes the longest. You cannot begin too early or too young, yet cannot expect a child to understand it at that age either. Unless teaching the mosaic principle begins in an entirely different situation.

This brings us back to one of my favourite subjects. Essentially, four important considerations in magical theory.

  • Spells affect whole objects, not individual parts. But is a latch a whole object, or a lock, or only the house or city? What makes an object a complete object?
  • You must know the object you shape, without needing to know how to manufacture it. You do need to know the concepts themselves.
  • Third, from the gift of tongues through the linguistic basis of described magical forms to the steps required in casting, everything points to magic's linguistic foundations.
  • Since the first edition, we have known that a hierarchical relationship exists between the material and mental planes. 

These are the stronger, system-related considerations. Weaker ones exist too. 

  • Why is gold harder to create than clay? — The wizard is not thinking in terms of Earth's material structure.
  • How can creating a substance be impossible while creating an alloy containing 9.99% of it is allowed? — What would happen if we "built from elements and compounds"?
  • When elemental magic extracts primal water from matter, the matter's structure nevertheless does not collapse.
  • Official sources also note that a material structure resembling Earth's would cause problems with Ethereality.

We know magical theory was not developed until the CODEX Grimoire, but also that there was no drastic change of worldview. That provides another source for creative thought, the idea, belonging to linguistically based magic. There are few contrary sources, unless we count the internal consensus of little sites collecting "new, even more overpowered classes found online here and there". On this basis, the mosaic principle assembles magical text from fragments, joining sections and substituting suitable concepts elsewhere.

Assembling texts and handling concepts are easy to learn if characters initially compose texts under proper guidance: sentences, short stories and fairy tales. Small logic games develop the reasoning involved. This gives us a realistic picture of how magical education can work despite one critical task's time requirements and the beginners' young age. There are few serious wizard schools. If lone masters and the like taught wizards on Ynev, many cities and regions would try persuading them to establish schools. Yet Ynev is not like that; there is no wizard school on every corner. Precisely because such a school needs infrastructure. 

Why? Tricky exercises commonly play a part. Give a character ten pictures to write a story about, one showing themselves, one a companion, and the rest a goblin, tower, sword, bow, large treasure pile, cart, barrel and cake, and their first thought is to write about themselves. Eventually, however, the story will concern not two particular people, but a girl and boy, a "peasant child" and beauty... Many similar variations are possible. This often needs personalised pictures, not merely supplied words, because words would prevent this play with images' meanings and the search for concepts: the characters would simply use the exact words given. 

That also means producing personalised pictures faster than future wizards and their masters use them up. Here we already begin to see why a wizard school necessarily operates with a larger staff.