Clarification of the term "Linear"
  <p>
    The term "linear" is very clear when applied to a
    mathematical function.  A function F is linear if and only
    if it obeys homogeneity and superposition:
    </p><p>
    Homogeneity:        F(cx) = cF(x)
    <br/>
    Superposition:      F(x+y) = F(x) + F(y)
    </p><p>
    In the context of what we have seen so far, the only
    elements that are linear as mathematical functions are
    resistors.  An independent voltage source or an independent
    current source is not a linear element.  (There are also
    linear dependent sources, linear capacitors and linear
    inductors, but we have not yet introduced them in our class.
    You will see them later.)
    </p><p>
    Formally, a circuit composed of only linear elements is a
    linear circuit.  When we add independent sources to a linear
    circuit as inputs, we get a circuit that is not linear
    because it has an offset: its v-i characteristic at a pair
    of exposed terminals may not pass through the origin.
    However, we can make a Thevenin or Norton equivalent model
    of such a circuit: the Thevenin resistance summarizes the
    effect of the linear elements and the Thevenin voltage
    summarizes the effect of the independent sources.
    </p><p>
    However the term "linear," when applied to an electrical
    circuit often takes on an informal meaning.  We often say
    that a circuit containing only linear elements and
    independent sources is a "linear circuit."  So, in the
    informal sense, a linear circuit is one where we can apply
    the Thevenin or Norton theorems to summarize the behavior at
    a pair of exposed terminals.
    </p><p>
    Sorry for the confusion of words &mdash; natural language is like
    that!
  </p>