Oersted Medal Lecture 2002: Reforming the Mathematical Language of Physics David Hestenes Arizona State University Reforming the Mathematical Language of Physics is the single most essential step toward simplifying and streamlining physics education

at all levels from high school through graduate school The relation between Teaching and Research is a perennial theme in academia and Oersted Lectures. PER puts the whole subject in a new light because it makes teaching itself a subject of research. The common denominator of T & R is learning! Learning by students Learning by scientists Without getting deeply into learning theory,

I want to supply you with a nontrivial example showing how integration of PER with scientific research can facilitate learning and understanding by both students and research physicists. Five Principles of Learning that have guided my own work in PER 1. Conceptual learning is a creative act. 2. Conceptual learning is systemic. 3. Conceptual learning is context dependent. 4. The quality of learning is critically dependent

on the quality of conceptual tools at the learners command. 5. Expert learning requires deliberate practice with critical feedback. Mathematical tools for introductory physics in 2D Vectors are primary tools for representing magnitude and direction But vector algebra cannot be used for reasoning with vectors because a b does not work in 2D Complex numbers are ideal for 2D rotations and trigonometry, but they are seldom used for lack of time

and generalizability to 3D A PER study at U. Maryland found that student use of vectors is best described as vector avoidance! Student learning is limited by almost exclusive reliance on weak coordinate methods This problem is not so much with the pedagogy as with the mathematical tools It is symptomatic of a larger problem with the math in physics! A Babylon of mathematical tongues contributes to fragmentation of knowledge

Babylon can be replaced by a single Geometric Algebra a unified mathematical language for the whole of physics ! To reform the mathematical language of physics, you need to start all over at the most elementary level. You need to relearn how to multiply vectors. My purpose today is to show you how and convince you that it is important! How to multiply vectors Multiplication in geometric algebra is nearly the same as in scalar algebra

Rules for the geometric product ab of vectors: (ab)c = a(bc) a(b + c) = ab + ac (b + c)a = ba + ca 2 2 a =a a associative left distributive right distributive

contraction = magnitude These are the basic grammar rules for GA, and they apply to vector spaces of any dimension. The power of GA derives from the simplicity of the grammar, the geometric meaning of the product ab. the way geometry links the algebra to physics Geometric Product ab implies two other products

with familiar geometric interpretations. Inner Product: Outer Product: a b = 12 (ab + ba) = b a a b = 12 (ab ba) = b a Bivector represents an oriented area by a Parallelogram rule:

(improves a b) Inner and outer products are parts of a single Geometric Product: ab = a

b The resulting object ab is a new entity with a different kind of geometric interpretation! + a b

Understanding the import of this formula: ab = a b

+ a b is the single most important step in unifying the mathematical language of physics. This formula integrates the concepts of vector

complex number quaternion spinor Lorentz transformation And much more! This lecture concentrates on how it integrates vectors and complex numbers into a powerful tool for 2D physics. Consider first the important special case of a unit bivector i It has two kinds of geometric interpretation! Object interpretation as an oriented area (additive) Can construct i from a pair of orthogonal unit vectors:

a b = 0

i a2 = b2 = 1 = a b

= ab = b ba

i2 = 1 So i oriented unit area for a plane a Operator interpretation as rotation by 90o (multiplicative) depicted as a directed arc b So i rotation by a right angle: 2

ai = b 2 2 Proof: (ab ) = ( ab )( ab ) = ( ba )( ab ) = a b = 1 Proof: ai = a(ab) = b a The operational interpretation of i generalizes to the concept of Rotor U , the entity produced by the geometric product ab of unit vectors with relative angle .

Rotor U is depicted as a directed arc on the unit circle. a2 = b2 = 1 Reversion: Uy = ba ab = U Defining sine and cosine functions from products of unit vectors a2 = b2 = 1

Defining sine and cosine functions from products of unit vectors a2 = b2 = 1 i = unit bivector i 2 = 1 Defining sine and cosine functions from products of unit vectors a2 = b2 = 1 i = unit bivector i 2 = 1

a b cos Defining sine and cosine functions from products of unit vectors a2 = b2 = 1 i = unit bivector i 2 = 1 a b cos a b i sin Defining sine and cosine functions from products of unit vectors

a2 = b2 = 1 i = unit bivector i 2 = 1 a b cos a b i sin Rotor: U = ab = a b + a b = cos + i sin = ei The concept of rotor generalizes to the concept of

complex number interpreted as a directed arc. z = U = ei = ab Reversion = complex conjugation z y = Uy = e i = ba Modulus zz y = 2 = (ab )(ba) = a 2 b 2 = z z = = ab This represention of complex numbers in a real GA is a special case of spinors for 3D.

2 z = Re z + i Im z = ab 1 Re z = ( z + z y ) = a b 2 1 i Im z = ( z zy ) = a b 2 Our development of GA to this point is sufficient to

formulate and solve any problem in 2D physics without resorting to coordinates. Of course, like any powerful tool, it takes some skill to apply it effectively. For example, every physicist knows that skillful use of complex numbers avoids decomposing them into real and imaginary parts whenever possible. Likewise, skillful use of the geometric product avoids decomposing it into inner and outer products. In the remainder of this lecture I demonstrate how rotor algebra facilitates the treatment of rotations in 2D & 3D. In particular, note the one-to-one correspondence

between algebraic operations and geometric depictions! Properties of rotors Rotor equivalence is like Vector equivalence of directed arcs of directed line segments

Properties of rotors Rotor equivalence is like Vector equivalence of directed arcs of directed line segments

Properties of rotors Rotor equivalence is like Vector equivalence of directed arcs of directed line segments Properties of rotors

Product of rotors U, U Addition of arcs UU =

U Properties of rotors Rotor-vector product = vector U, v Uv = u

Rotor products composition of rotations in 3D U1 Rotor products in 3D U1 , U2 Rotor products in 3D U1 , U2 U2U1 Rotor products in 3D

U1 , U2 U2U1 = (bc)(ca) Rotor products in 3D U1 , U2 U2U1 = (bc)(ca) = ba = U3 U2 U1 = U3 Noncommutativity of Rotations U2 (U1) = U2U1 U1 (U2) = U1U2

Noncommutativity of Rotations U2 (U1) Noncommutativity of Rotations U2 (U1) = U2U1 Noncommutativity of Rotations U2 (U1) = U2U1 U1 (U2 Noncommutativity of Rotations

U2 (U1) = U2U1 U1 (U2) Noncommutativity of Rotations U2 (U1) = U2U1 U1 (U2) = U1U2 What have we learned so far? Rules for multiplying vectors that apply to vector spaces of any dimension. Geometric meaning of the geometric product and its component parts in

ab = a b + a b Integration of complex numbers with vectors and interpretation as directed arcs. How rotor algebra clarifies and facilitates the treatment of rotations in 2D and 3D. What next? A greatly expanded written version of this lecture, to be published in the AJP, demonstrates how GA integrates and simplifies

classical, relativistic and quantum physics, develops GA to the point where it is ready to incorporate into the physics curriculum at all levels. I will conduct a Workshop on GA in physics at the summer AAPT meeting. For those who cant wait, A thorough introduction to 3D GA with applications in my book New Foundations for Classical Mechanics Many papers and links to other web sites @ http:\\ modelingnts.la.asu.edu A challenge to PER and the physics community!

Critically examine the following claims: GA provides a unified language for the whole physics that is conceptually and computationally superior to alternative math systems in every application domain. GA can enhance student understanding and accelerate student learning of physics. GA is ready to incorporate into the physics curriculum. Research on the design and use of mathematical tools is equally important for instruction and for theoretical physics. Scientific Research

Learning Theory How people learn Modeling Theory How science works Theory of Instruction Teaching Practice What follows? The coordinate-free representation of rotations

by rotors and directed arcs generalizes to Rotations in 3D, Lorentz transformations in spacetime, Real spinors in quantum mechanics, where the unit imaginary appears as a unit bivector i related to spin!