octonion in nLab
- ️Invalid Date
Context
Arithmetic
- natural number, integer number, rational number, real number, irrational number, complex number, quaternion, octonion, adic number, cardinal number, ordinal number, surreal number
-
transfinite arithmetic, cardinal arithmetic, ordinal arithmetic
-
prime field, p-adic integer, p-adic rational number, p-adic complex number
arithmetic geometry, function field analogy
Algebra
- algebra, higher algebra
- universal algebra
- monoid, semigroup, quasigroup
- nonassociative algebra
- associative unital algebra
- commutative algebra
- Lie algebra, Jordan algebra
- Leibniz algebra, pre-Lie algebra
- Poisson algebra, Frobenius algebra
- lattice, frame, quantale
- Boolean ring, Heyting algebra
- commutator, center
- monad, comonad
- distributive law
Group theory
- group, normal subgroup
- action, Cayley's theorem
- centralizer, normalizer
- abelian group, cyclic group
- group extension, Galois extension
- algebraic group, formal group
- Lie group, quantum group
Ring theory
Module theory
Gebras
Exceptional structures
exceptional structures, exceptional isomorphisms
Examples
-
exceptional finite rotation groups:
-
and Kac-Moody groups:
-
exceptional Jordan superalgebra, K 10K_10
Interrelations
Applications
Philosophy
Octonions
Idea
The octonions or Cayley numbers (Cayley 1845) form a non-associative real star-algebra 𝕆\mathbb{O} similar to the complex numbers and the quaternions but with seven imaginary units adjoined.
The octonions arise from the quaternions in analogy – namely: by the Dickson double construction (Dickson 1919, (6)) – of how the quaternions arise from the complex numbers, and the complex numbers from the real numbers. These are precisely the normed division algebras over the real numbers, the octonions being the largest of the four. While the further Dickson double of the quaternions exists, called the sedenions, it is no longer a normed division algebra.
In continuation of how the complex numbers and quaternions control spin groups, real spin representations and supersymmetry up to dimension 7, the octonions control these up to the maximal dimension 11:
exceptional spinors and real normed division algebras
Lorentzian spacetime dimension | AA\phantom{AA}spin group | normed division algebra | \,\, brane scan entry |
---|---|---|---|
3=2+13 = 2+1 | Spin(2,1)≃SL(2,ℝ)Spin(2,1) \simeq SL(2,\mathbb{R}) | A\phantom{A} ℝ\mathbb{R} the real numbers | super 1-brane in 3d |
4=3+14 = 3+1 | Spin(3,1)≃SL(2,ℂ)Spin(3,1) \simeq SL(2, \mathbb{C}) | A\phantom{A} ℂ\mathbb{C} the complex numbers | super 2-brane in 4d |
6=5+16 = 5+1 | Spin(5,1)≃Spin(5,1) \simeq SL(2,H) | A\phantom{A} ℍ\mathbb{H} the quaternions | little string |
10=9+110 = 9+1 | Spin(9,1) ≃{\simeq} “SL(2,O)” | A\phantom{A} 𝕆\mathbb{O} the octonions | heterotic/type II string |
For more on this see at supersymmetry and division algebras.
Generally, the algebra of octonions shows up, in one way or another, behind most, if not all, exceptional structures in group theory, Lie theory and differential geometry. See also at universal exceptionalism for more on this.
Definition
The following definition is in the style of Dickson 1919, Baez 02, second half of Section 2.2:
Definition
The octonions 𝕆\mathbb{O} are the elements of the non-associative star-algebra over the real numbers which is the Cayley-Dickson double of the star-algebra of quaternions ℍ\mathbb{H} (with (−)¯\overline{(-)} denoting the conjugation-operation).
This means (see there) that if i,j,k∈ℍi, j, k \in \mathbb{H} denote an orthonormal basis of imaginary unit-quaternions
i 2=j 2=k 2=−1 ij=k,ji=−kand cyclic \begin{aligned} & i^2 = j^2 = k^2 = -1 \\ & i j = k, \, j i = - k \;\;\;\text{and cyclic} \end{aligned}
then the algebra 𝕆\mathbb{O} of octonions is generated from these i,j,ki, j , k and from one more element ℓ\ell, subject to these relations:
ℓ 2=−1,AAAℓ¯=−ℓ \begin{aligned} \ell^2 \;=\; -1 \,, \phantom{AAA} \overline{\ell} \;=\; - \ell \end{aligned}
and
q(ℓq′)=ℓ(q¯q′) (qℓ)q′=(qq¯′)ℓ (ℓq)(q′ℓ)=−qq′¯ \begin{aligned} q (\ell q') = \ell (\overline{q} q') \\ (q \ell) q' = (q \overline{q}') \ell \\ (\ell q) (q' \ell) = - \overline{q q'} \end{aligned}
for all quaternions q 1,q 2∈ℍq_1, q_2 \in \mathbb{H}.
This gives the multiplication table on the right, where any two consecutive arrows a→b→ca \to b \to c mean that ab=ca b = c, ca=bc a = b, bc=ab c = a and ba=−cb a = -c.
Example
The following computation shows the operation of consecutive left multiplication by the generators e 4\mathrm{e}_4, e 5\mathrm{e}_5, e 6\mathrm{e}_6 e 7\mathrm{e}_7 (according to Def. ) on any octonion x=q+pℓx = q + p \ell (q,p∈ℍq,p \in \mathbb{H}) is by reversal of the sign of the ℓ\ell-component, hence has as fixed linear subspace the quaternions (see HSS 18, Lemma 4.13 for application of this fact to M-branes):
e 4(e 5(e 6(e 7x))) =ℓ((iℓ)((jℓ)((kℓ)x))) =ℓ((iℓ)((jℓ)((kx¯)ℓ))) =ℓ((iℓ)((xk)j)) =ℓ((i(j(kx¯))ℓ) =((x⏟q+pℓk)j)i =qkji+(((qℓ)k)j)i =qkji⏟=1−(pkji⏟=1)ℓ =q−pℓ. \begin{aligned} \mathrm{e}_4 \Big( \mathrm{e}_5 \big( \mathrm{e}_6 (\mathrm{e}_7 x) \big) \Big) & = \ell \bigg( (i \ell) \Big( (j \ell) \big( (k \ell) x \big) \Big) \bigg) \\ & = \ell \bigg( (i \ell) \Big( (j \ell) \big( (k \overline{x}) \ell \big) \Big) \bigg) \\ & = \ell \Big( (i \ell) \big( (x k) j \big) \Big) \\ & = \ell \bigg( \Big( i \big( j (k \overline{x} \big) \Big) \ell \bigg) \\ & = \big( ( \underset{ \mathclap{ q + p \ell } }{ \underbrace{ x } } k) j \big) i \\ & = q k j i + \Big( \big( (q \ell) k \big) j \Big) i \\ & = q \underset{ = 1 }{ \underbrace{ k j i } } - (p \underset{ = 1 }{ \underbrace{ k j i } } ) \ell \\ & = q - p \ell \,. \end{aligned}
Of course the labels of the generators is not fixed. Here is another version:
Definition
The octonions 𝕆\mathbb{O} is the nonassociative algebra over the real numbers which is generated from seven generators {e 1,⋯,e 7}\{e_1, \cdots, e_7\} subject to the relations
-
for all ii
e i 2=−1e_i^2 = -1
-
for e i→e j→e ke_i \to e_j \to e_k an edge or circle in the following diagram (a labeled version of the Fano plane) the relations
-
e ie j=e ke_i e_j = e_k
-
e je i=−e ke_j e_i = -e_k
-
\,
This becomes a star-algebra with star involution
(1)(−)¯:𝕆⟶𝕆 \overline{(-)} \;\colon\; \mathbb{O} \longrightarrow \mathbb{O}
which is the antihomomorphism ab¯=b¯a¯\overline{a b} = \overline{b} \overline{a} that is given on the above generators by
e i¯≔−e iAAAAi∈{1,⋯,7}. \overline{e_i} \coloneqq - e_i \phantom{AAAA} i \in \{1, \cdots, 7\} \,.
Example
The product of all the generators with each other, bracketed to the right, is
e 7(e 6(e 5(e 4(e 3(e 2(e 11))))))=+1 e_7 (e_6 (e_5 (e_4 (e_3 ( e_2 (e_1 1 )))))) \;=\; + 1
Proof
By iteratively using the multiplication table in def. we compute as follows:
e 7(e 6(e 5(e 4(e 3(e 2e 1⏟−e 4))))) =−e 7(e 6(e 5(e 4(e 3e 4⏟e 6)))) =−e 7(e 6(e 5(e 4e 6⏟e 3))) =−e 7(e 6(e 5e 3⏟−e 2)) =+e 7(e 6e 2⏟−e 7) =−e 7e 7⏟=−1 =+1 \begin{aligned} & e_7 (e_6 (e_5 (e_4 (e_3 (\underset{-e_4}{\underbrace{e_2 e_1}}))))) \\ & = - e_7 (e_6 (e_5 (e_4 (\underset{e_6}{\underbrace{e_3 e_4}})))) \\ & = - e_7 (e_6 (e_5 (\underset{e_3}{\underbrace{e_4 e_6}}))) \\ & = - e_7 (e_6 (\underset{-e_2}{\underbrace{e_5 e_3}})) \\ & = + e_7 (\underset{-e_7}{\underbrace{e_6 e_2}}) \\ & = - \underset{= -1}{\underbrace{e_7 e_7}} \\ & = + 1 \end{aligned}
Definition
(real and imaginary octonions)
As for the complex numbers one says that
-
an imaginary octonion is an a∈𝕆a \in \mathbb{O} shuch that under the star involution (1) it is sent to its negative:
a¯=−a \overline{a} = -a
-
a real octonions is an a∈𝒪a \in \mathcal{O} shuch that under the star involution (1) it is sent to itself
a¯=a \overline{a} = a
Accordingly every octonion decomposes into a real part and an imaginary part:
Re(a)≔12(a+a¯)AAIm(a)≔12(a−a¯). Re(a) \coloneqq \tfrac{1}{2}(a + \overline{a}) \phantom{AA} Im(a) \coloneqq \tfrac{1}{2}(a - \overline{a}) \,.
Properties
General
Proof
By linearity it is sufficient to check this on generators. So let e i→e j→e ke_i \to e_j \to e_k be a circle or a cyclic permutation of an edge in the Fano plane as in Def. . Then by definition of the octonion multiplication we have
(e ie j)e j =e ke j =−e je k =−e i =e i(e je j) \begin{aligned} (e_i e_j) e_j &= e_k e_j \\ &= - e_j e_k \\ & = -e_i \\ & = e_i (e_j e_j) \end{aligned}
and similarly
(e ie i)e j =−e j =−e ke i =e ie k =e i(e ie j). \begin{aligned} (e_i e_i ) e_j &= - e_j \\ &= - e_k e_i \\ &= e_i e_k \\ &= e_i (e_i e_j) \end{aligned} \,.
Automorphisms
See also at normed division algebra – automorphism
Left multiplication by imaginary octonions
Definition
Given any octonion oo, then the operation of left multiplication by oo
𝕆 ⟶L o 𝕆 a ↦ oa \array{ \mathbb{O} &\overset{L_o}{\longrightarrow}& \mathbb{O} \\ a &\mapsto& o a }
is an ℝ−\mathbb{R}-linear map. Under composition of linear maps, this defines an associative monoid acting linearly on 𝕆\mathbb{O}.
Proposition
(Clifford action of imaginary octonions)
Consider the Clifford algebra
Cl(Im(𝕆),−|−| 2) Cl(Im(\mathbb{O}), -{\vert -\vert}^2)
on the underlying real vector space of that of the imaginary octonions (Def. ) regarded as an inner product space via the quadratic form given by the negative square norm.
Then the operation of left multiplication on 𝕆\mathbb{O} (def. ) induces a representation of this Clifford algebra on ℝ 8≃ ℝ𝕆\mathbb{R}^8 \simeq_{\mathbb{R}} \mathbb{O}.
Proof
By alternativity (Prop. ) we have for every v∈Im(𝕆)v \in Im(\mathbb{O}) and every a∈𝕆a \in \mathbb{O}
L vL v(a) ≔v(va) =(vv)a =−|v| 2a =L −|v| 2(a) \begin{aligned} L_v L_v (a) & \coloneqq v (v a) \\ & = (v v) a \\ & = - {\vert v\vert}^2 a \\ & = L_{- {\vert v\vert}^2} (a) \end{aligned}
Proposition
(consecutive left action by imaginary generators is unity)
The consecutive left multiplication action (Def. ) by all the imaginary octonion generators e ie_i (Def. ) is ±\pm the identity function on the octonions. Specifically, if one acts in increasing order of the labels in Def. , then it is +1:
L e 7L e 6L e 5L e 4L e 3L e 2L e 1=+Id 𝕆 L_{e_7} L_{e_6} L_{e_5} L_{e_4} L_{e_3} L_{e_2} L_{e_1} \;=\; + Id_{\mathbb{O}}
Proof
All the generators e ie_i are imaginary octonions (Def. ). By Prop. their left action on 𝕆\mathbb{O} represents a Clifford algebra-action of Cl(Im(𝕆),−|−| 2)≃Cl 0,7Cl(Im(\mathbb{O}), -{\vert-\vert}^2) \simeq Cl_{0,7} on ℝ 8≃ ℝ𝕆\mathbb{R}^{8} \simeq_{\mathbb{R}} \mathbb{O}.
By the classification of real Clifford algebras, Cl 0,7Cl_{0,7} has, up to isomorphism, two different irreducible modules. Their underlying vector space is ℝ 8\mathbb{R}^8 in both cases, and so the left action of imaginary octonions we have must be one of the two. The two irreps may be distinguished by the action of the “volume element” Γ 7Γ 6⋯Γ 1\Gamma_7 \Gamma_6 \cdots \Gamma_1: On one of the two it acts as the identity, on the other as minus the identity.
Hence we may check the remaining sign by acting on any one octonion, for instance on the unit 1∈𝕆1 \in \mathbb{O}. Then the claim follows with the computation in Example :
L e 7L e 6L e 5L e 4L e 3L e 2L e 1(1) =e 7(e 6(e 5(e 4(e 3(e 2(e 11)))))) =+1. \begin{aligned} L_{e_7} L_{e_6} L_{e_5} L_{e_4} L_{e_3} L_{e_2} L_{e_1} (1) & = e_7 (e_6 (e_5 (e_4 (e_3 ( e_2 (e_1 1 )))))) \\ & = + 1 \,. \end{aligned}
Basic triples
Definition
A special triple or basic triple is a triple (e 1,e 2,e 3)∈𝕆 3(e_1, e_2, e_3) \in \mathbb{O}^3 of three octonions such that
-
e i 2=−1e_i^2 = -1
-
e ie j=−e je ie_i e_j = - e_j e_i for i≠ji \ne j
-
e i(e je k)=−(e je k)e ie_i (e_j e_k) = - (e_j e_k) e_i for i,j,ki,j,k pairwise distinct.
(e.g. Baez 02, 4.1)
Relation to quaternions
Proposition
Let
ℍ=⟨1,i,j,k⟩ \mathbb{H} = \langle 1, i, j, k\rangle
be the quaternions equipped with canonical basis elements, and let
(2)𝕆=ℍ⊕ℓℍ \mathbb{O} = \mathbb{H} \oplus \ell \mathbb{H}
be the octonions equipped with the linear basis induced by the Cayley-Dickson construction (via this def.).
Then the linear map
L ℓL ℓiL ℓjL ℓk:𝕆⟶𝕆 L_{\ell} L_{\ell i} L_{\ell j} L_{\ell k} \;\colon\; \mathbb{O} \longrightarrow \mathbb{O}
is an involution whose +1 eigenspace is ℓℍ\ell \mathbb{H} and whose -1 eigenspace is ℍ\mathbb{H}, under the above identification (2).
(Here L (−)L_{(-)} denotes the linear map on 𝕆\mathbb{O} given by left multiplication in 𝕆\mathbb{O}.)
Proof
We use the Cayley-Dickson relations (this def.)
a(ℓb)=ℓ(a¯b),AAa(ℓb)=(ab¯)ℓ,AA(ℓa)(bℓ −1)=ab¯ a (\ell b) = \ell (\overline{a} b) \,, \phantom{AA} a(\ell b) = (a \overline{b}) \ell \,, \phantom{AA} (\ell a) (b \ell^{-1}) = \overline{a b}
that hold in 𝕆\mathbb{O} for all a,b∈ℍa,b \in \mathbb{H}, as well as
ℓe=−eℓ \ell e = - e \ell
for all imaginary elements e∈ℍe \in \mathbb{H}.
With this we compute
L ℓL ℓiL ℓjL ℓk(a) =ℓ((ℓi)((ℓj)((ℓk)a))) =−ℓ((ℓi)((ℓj)((kℓ)a))) =−ℓ((ℓi)((ℓj)((ka¯)ℓ))) =+ℓ((ℓi)((ℓj)((ka¯)ℓ −1))) =+ℓ((ℓi)(j(ka¯)¯)) =+ℓ((ℓi)(ia¯¯)) =−ℓ((iℓ)(ia¯¯)) =−ℓ((iia¯)ℓ) =+ℓ(a¯ℓ) =−ℓ(a¯ℓ −1) =−a \begin{aligned} & L_{\ell} L_{\ell i} L_{\ell j} L_{\ell k} ( a) \\ & = \ell( (\ell i) ( (\ell j) ( (\ell k) a ) ) ) \\ & = - \ell( (\ell i) ( (\ell j) ( (k \ell) a ) ) ) \\ & = - \ell( (\ell i) ( (\ell j) ( (k \overline{a}) \ell ) ) ) \\ & = + \ell( (\ell i) ( (\ell j) ( (k \overline{a}) \ell^{-1} ) ) ) \\ & = + \ell( (\ell i) ( \overline{j (k \overline{a})} ) ) \\ & = + \ell( (\ell i) ( \overline{i \overline{a}} ) ) \\ & = - \ell( (i \ell) ( \overline{i \overline{a}} ) ) \\ & = - \ell( ( i i \overline{a} ) \ell ) \\ & = + \ell( \overline{a} \ell ) \\ & = - \ell( \overline{a} \ell^{-1} ) \\ & = - a \end{aligned}
and
L ℓL ℓiL ℓjL ℓk(ℓa) =ℓ((ℓi)((ℓj)((ℓk)(ℓa)))) =ℓ((ℓi)((ℓj)((ℓk)(a¯ℓ)))) =−ℓ((ℓi)((ℓj)(ka¯¯))) =+ℓ((ℓi)((jℓ)(ka¯¯))) =+ℓ((ℓi)((jka¯)ℓ)) =−ℓ((ℓi)((jka¯)ℓ −1)) =−ℓ(ijka¯¯) =+ℓ(a¯¯) =+ℓa \begin{aligned} & L_{\ell} L_{\ell i} L_{\ell j} L_{\ell k} ( \ell a) \\ & = \ell( (\ell i) ( (\ell j) ( (\ell k) (\ell a) ) ) ) \\ & = \ell( (\ell i) ( (\ell j) ( (\ell k) ( \overline{a} \ell) ) ) ) \\ & = - \ell( (\ell i) ( (\ell j) ( \overline{k {\overline{a}}} ) ) ) \\ & = + \ell( (\ell i) ( (j \ell) ( \overline{k {\overline{a}}} ) ) ) \\ & = + \ell( (\ell i) ( (j k {\overline{a}}) \ell ) ) \\ & = - \ell( (\ell i) ( (j k {\overline{a}}) \ell^{-1} ) ) \\ & = - \ell( \overline{ i j k {\overline{a}} } ) \\ & = + \ell( \overline{ {\overline{a}} } ) \\ & = + \ell a \end{aligned}
exceptional spinors and real normed division algebras
Lorentzian spacetime dimension | AA\phantom{AA}spin group | normed division algebra | \,\, brane scan entry |
---|---|---|---|
3=2+13 = 2+1 | Spin(2,1)≃SL(2,ℝ)Spin(2,1) \simeq SL(2,\mathbb{R}) | A\phantom{A} ℝ\mathbb{R} the real numbers | super 1-brane in 3d |
4=3+14 = 3+1 | Spin(3,1)≃SL(2,ℂ)Spin(3,1) \simeq SL(2, \mathbb{C}) | A\phantom{A} ℂ\mathbb{C} the complex numbers | super 2-brane in 4d |
6=5+16 = 5+1 | Spin(5,1)≃Spin(5,1) \simeq SL(2,H) | A\phantom{A} ℍ\mathbb{H} the quaternions | little string |
10=9+110 = 9+1 | Spin(9,1) ≃{\simeq} “SL(2,O)” | A\phantom{A} 𝕆\mathbb{O} the octonions | heterotic/type II string |
References
General
The definition is originally due to
- Arthur Cayley, On certain results relating to quaternions, The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science Series 3 Volume 26, 1845 - Issue 171 (doi:10.1080/14786444508562684)
The formulation as the Dickson double construction is due to
- Leonard Dickson, On Quaternions and Their Generalization and the History of the Eight Square Theorem,
Annals of Mathematics, Second Series, Vol. 20, No. 3 (Mar., 1919), pp. 155-171 (jstor:1967865)
Review:
- John Baez, The Octonions, Bull. Amer. Math. Soc. 39 (2002), 145-205. (web, pdf doi:10.1090/S0273-0979-01-00934-X)
Textbook accounts:
-
Tonny Springer, Ferdinand Veldkamp, Octonions, Jordan Algebras, and Exceptional Groups, Springer Monographs in Mathematics, 2000 (doi:10.1007/978-3-662-12622-6)
-
Tevian Dray, Corinne Manogue, The Geomety of Octonions, World Scientific 2015 (doi:10.1142/8456)
The concept of “special triples” or (“basic triples”) used above seems to go back to
- George Whitehead, appendix A in Homotopy Theory, MIT press 1971
Relation to the Leech lattice:
- Robert A. Wilson, Octonions and the Leech lattice, Journal of Algebra
Volume 322, Issue 6, 15 September 2009, Pages 2186-2190, (pdf, slides)
Relation to 10d/11d spin geometry
Application of octonion-algebra to analysis of spin representations and spin geometry specifically in 11d (for general discussion in other dimensions see at supersymmetry and division algebras):
-
John Huerta, Hisham Sati, Urs Schreiber, Real ADE-equivariant (co)homotopy and Super M-branes, Comm. Math. Phys. 371: 425. (2019) (arXiv:1805.05987)
-
Hisham Sati, Urs Schreiber, Super-exceptional M5-brane model – Emergence of SU(2)SU(2)-flavor sector, Journal of Geometry and Physics 170 (2021) 104349 [arXiv:2006.00012, doi:10.1016/j.geomphys.2021.104349]
Last revised on August 21, 2024 at 02:06:16. See the history of this page for a list of all contributions to it.