Maths - Duality

One of the simplest and most fundamental dualities is the duality between points and lines in 2D space. Expressions that involve points and lines can be rewritten in a way that points and lines are interchanged, for instance:

In 3 dimensions the dualities change, in this case there is a duality between points and planes, in 'n' dimensions we can predict the dualities because they have a binomial structure. This is reflected in the structure of Geometric Algebra, as explained here, which makes Geometric Algebra useful when we want to transfer geometry and physics between a different number of dimensions.

It is not only in geometry and algebra that these dualities occur, many of the physics quantities have corresponding dualities, here are some examples:

Field First Dual Second Dual Example Applies in switch





  any dimensions ? gram matrix
geometric algebra vector


directed area

  3 dimensions multiply by pseudo-scalar
calculus differentiation integration electrostatic and magnetic fields any dimensions  
calculus normal space tangent space      
field theory scalar field vector field electro-magnetic fields, kirchoff's law 3 dimensions diff or integrate

inner product

(dot product)

outer product  

any dimensions

3 for cross product

algebra a 1/a   any dimensions invert
orthogonal matrix rows columns   orthogonal coordinates transpose
transform sandwich products q P q-1 q-1 P q   any dimensions reverse transform
platonic solids



  3 dimensions swap faces and verticies.

From some points of view, for example when studying tensors, it almost seems like these dualitys are all the expression of the same correspondence. its as if we can put all the things in the 'first dual' column together and they form a consistent model of the world and then we can put all the things in the 'second dual' column together and they form a mirror image model of the original world.

However I have to restrain myself from taking it too far. There are lots of limits to this, for example some of these things only work in 3 dimensions (in 4 dimensions there may be a different set of dualities, for example vectors and trivectors). There may also be other constraints, some things may depend on orthogonal coordinates.

Perhaps the above model is too simplistic in that i have mixed together elements like scalar and vector fields together with operators which switch between them. It might be better to draw this in a two dimensional way like a morphism?


There are another set of dualities which I'm tempted to add to the above (perhaps that's taking things a bit too far, I don't know?):

Field First Dual Second Dual Example Applies in switch
logic and gates or gates De Morgan duality   not gate
probability permutations combinations     1-prob
symmetry right handed left image mirror image   mirror image
parity even odd      

There are arguments that I could make to add these, for instance, both seem to have a mirror image property and the binomial triangle seems to be involved in both?

A lot more speculative but I get the impression that there is a duality between translational and rotational quantities, as suggested on this page, such as:

Field First Dual Second Dual Example Applies in switch
transforms translation rotation      
coordinates rectangular polar      
quantum mechanics particles waveform      

Vector, Covector and Bivector

Covectors come from tensor algebra whereas bivectors come from Clifford/geometric algebras however despite their different origins, if we restrict ourselves purely to 3 dimensional euclidean space, then the two ideas turn out to represent the same thing as explained on this page.

Contravarient and Covariant

When defining vectors we use the linear combination of basis vectors. Since the physical vector is fixed, then, if we change the basis vectors then the vector components will automatically change in the inverse way.

Scalar Field and Vector Field

I always thought that scalar and vector fields were completely different physical entities, for instance,

However we can always convert from one type to another type by differentiation and integration so, for example, temperature gradient would be a vector field.

The other duality is combining two vector fields by using the dot product to give a scalar field (or a vector fields with its dual by using the dot product to give a scalar field).

Tangent Space

In addition to the scalar and vector fields above we have a duality between the vector tangent and normal spaces.

Differentiation and Integration

Or the vector calculus such as:

Inner Product and Outer Product

Rows and Columns

When using a matrix transform based on orthogonal coordinates then transposing the rows and columns inverts the transform.

Geometric Algebra Transforms

Transforms in geometric algebra are represented by sandwich products q P q-1 and q-1 P q which are inverses of each other. That is if we apply q P q-1 to a point and then apply q-1 P q to the result we get back to the original point.

metadata block
see also:
Correspondence about this page

Book Shop - Further reading.

Where I can, I have put links to Amazon for books that are relevant to the subject, click on the appropriate country flag to get more details of the book or to buy it from them.

cover The MathML Handbook - for people interested in working with mathematics on the web.

Other Math Books

Terminology and Notation

Specific to this page here:


This site may have errors. Don't use for critical systems.

Copyright (c) 1998-2017 Martin John Baker - All rights reserved - privacy policy.