In a few situations, I found myself being asked by younger students why the concepts of a basis was important.

First, in the concept of linear spaces, it's easy to explain that having a basis allows us to describe the spaces with a relatively small set and makes working with linear transformation a much easier task (there are many more reasons, but this was an introductory linear algebra course).

When I was asked the same question in the context of propositional logic (a basis here is a maximally independent set of formulas), I found it a bit more difficult to motivate the definition.

In general, when asked "What is the importance of a basis?", what do you reply?

  • $\begingroup$ The most basic question in linear algebra is "describe the general solution to $Ax=b$." Doing that amounts to writing $x=x_0+\sum_{i=1}^n c_i k_i$, where $c_i$ are arbitrary scalars and $k_i$ make a basis for the kernel of $A$. Each choice of the $c_i$ gives a different solution. So the meaningful information here is the particular solution $x_0$ and the basis $k_i$ for the kernel. They already know this fact before they get to vector spaces in the first place; it is just formal language for something they already understand. $\endgroup$ – Ian Apr 9 '16 at 21:18
  • $\begingroup$ I think putting everything in this framework will help them, especially if they cannot easily understand why a statement like "a linear transformation is determined by its action on a basis" is useful. $\endgroup$ – Ian Apr 9 '16 at 21:20

In every context, a basis is essentially a (not the) smallest set you need to understand, in order to understand the whole object in question.

For example, a linear transformation is determined by its action on a basis. Similarly, a complete theory is determined by which elements of a (propositional) basis it contains.

The virtue of the word "smallest" in the first paragraph is that, roughly speaking, every behavior over a basis is possible. For example, if $\{b_i\}$ ($i\in I$) is a basis for $V$ and $w_i$ $(i\in I)$ are vectors in $W$, then there is a linear transformation $T$ sending $b_i$ to $w_i$ (and by the above paragraph, it's unique). Similarly, given a propositional basis $p_i$ $(i\in I$) and a map $t: I\rightarrow \{\top, \bot\}$, there is a complete theory $T$ such that $p_i\in T\iff t(i)=\top$ (and by the above, it's unique).

These two aspects of bases (determination and freedom) mean that they're invaluable for understanding the things that act (in a very loose sense) on them - linear transformations, or complete theories, or etc.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.