In category theory, a Lawvere theory (named after American mathematician William Lawvere) is a category that can be considered a categorical counterpart of the notion of an equational theory. Intuitively, it is a categorical generalization of algebraic structures (e.g., a group or a ring), where there exists a "generic" object and all objects are isomorphic to an integer power of
x
{\displaystyle x}
, representing the inputs for the
n
{\displaystyle n}
-ary operations on
x
{\displaystyle x}
(i.e., of the form
x
n
↦
x
{\displaystyle x^{n}\mapsto x}
, starting from the fact that
x
:=
x
1
{\displaystyle x:=x^{1}}
and so
x
∘
x
:=
x
1
∘
x
1
:=
x
1
x
1
:=
x
2
{\displaystyle x\circ x:=x^{1}\circ x^{1}:=x^{1}x^{1}:=x^{2}}
; trivially generalizing inductively, we get the rest of the objects) where the operations come from the algebraic structure at hand (e.g., addition and/or multiplication).
The Lawvere theory of groups has as its generic object an underlying placeholder
x
{\displaystyle x}
where the other objects are the inputs for
n
{\displaystyle n}
-ary operations from those integer powers of
x
{\displaystyle x}
of the form (
x
n
{\displaystyle x^{n}}
) back to
x
{\displaystyle x}
, where a model, a finite-product preserving functor, from this theory into a target category
C
{\displaystyle C}
such as the category of sets or topological spaces, would map the abstract theory
x
{\displaystyle x}
onto the category to create a concrete, "combined," group-based structure. This model would provide a set with group structure (a group) or a topological space with group structure (a topological group), supplying appropriate names to the generic object
x
{\displaystyle x}
and its mappings (
n
{\displaystyle n}
-ary operations) according to whatever the theory and model at play are; in the model of sets for the Lawvere theory of groups, the generic object is a group and its mappings are group operations.
Definition
Let
ℵ
0
{\displaystyle \aleph _{0}}
be a skeleton of the category FinSet of finite sets and functions. Formally, a Lawvere theory consists of a small category
L
{\displaystyle L}
with (strictly associative) finite products and a strict identity-on-objects functor
I
:
ℵ
0
op
→
L
{\displaystyle I:\aleph _{0}^{\text{op}}\rightarrow L}
preserving finite products.
A model of a Lawvere theory in a category
C
{\displaystyle C}
with finite products is a finite-product preserving functor
M
:
L
→
C
{\displaystyle M:L\rightarrow C}
. A morphism of models
h
:
M
→
N
{\displaystyle h:M\rightarrow N}
where
M
{\displaystyle M}
and
N
{\displaystyle N}
are models of
L
{\displaystyle L}
is a natural transformation of functors.
Model examples
Some examples of models of the Lawvere theory of groups (i.e.,
L
{\displaystyle L}
=
L
a
w
G
r
p
{\displaystyle \mathbf {LawGrp} }
):
-
M
:
L
→
S
e
t
{\displaystyle M:L\rightarrow \mathbf {Set} }
( M {\displaystyle M}
is a classical group)
-
M
:
L
→
T
o
p
{\displaystyle M:L\rightarrow \mathbf {Top} }
( M {\displaystyle M}
is a topological group)
-
M
:
L
→
M
a
n
{\displaystyle M:L\rightarrow \mathbf {Man} }
( M {\displaystyle M}
is a Lie group).
Some examples of models of the Lawvere theory of rings (i.e.,
L
{\displaystyle L}
=
L
a
w
R
i
n
g
{\displaystyle \mathbf {LawRing} }
):
-
M
:
L
→
S
e
t
{\displaystyle M:L\rightarrow \mathbf {Set} }
( M {\displaystyle M}
is a classical ring)
-
M
:
L
→
T
o
p
{\displaystyle M:L\rightarrow \mathbf {Top} }
( M {\displaystyle M}
is a topological ring)
-
M
:
L
→
M
a
n
{\displaystyle M:L\rightarrow \mathbf {Man} }
( M {\displaystyle M}
is a smooth ring).
Category of Lawvere theories
A map between Lawvere theories
(
L
,
I
)
{\displaystyle (L,I)}
and
(
L
′
,
I
′
)
{\displaystyle (L',I')}
is a finite-product preserving functor that commutes with
I
{\displaystyle I}
and
I
′
{\displaystyle I'}
. Such a map is commonly seen as an interpretation of
(
L
,
I
)
{\displaystyle (L,I)}
in
(
L
′
,
I
′
)
{\displaystyle (L',I')}
.
More formally, a map
K
:
(
L
,
I
)
→
(
L
′
,
I
′
)
{\displaystyle K:(L,I)\rightarrow (L',I')}
, such that
K
∘
I
=
I
′
{\displaystyle K\circ I=I'}
, where
I
(
n
)
=
x
L
n
{\displaystyle I(n)=x_{L}^{n}}
and
(
K
∘
I
)
(
n
)
=
I
′
(
n
)
=
x
L
′
n
{\displaystyle (K\circ I)(n)=I'(n)=x_{L'}^{n}}
.
Lawvere theories together with maps between them form the category
L
a
w
{\displaystyle \mathbf {Law} }
.
Variations
Variations include multisorted (or multityped) Lawvere theory, infinitary Lawvere theory, and finite-product theory.[1]
See also
Notes
- Lawvere theory at the nLab
References
- Hyland, Martin; Power, John (2007), "The Category Theoretic Understanding of Universal Algebra: Lawvere Theories and Monads" (PDF), Electronic Notes in Theoretical Computer Science, 172 (Computation, Meaning, and Logic: Articles dedicated to Gordon Plotkin): 437–458, CiteSeerX 10.1.1.158.5440, doi:10.1016/j.entcs.2007.02.019
- Lawvere, William F. (1963), "Functorial Semantics of Algebraic Theories", PhD Thesis, vol. 50, no. 5, Columbia University, pp. 869–872, Bibcode:1963PNAS...50..869L, doi:10.1073/pnas.50.5.869, PMC 221940, PMID 16591125
Further reading
- Power, John (1999). "Enriched Lawvere theories" (PDF). Theory and Applications of Categories. 06: 83–93. doi:10.70930/tac/soye1d6v (inactive 13 April 2026).
{{cite journal}}: CS1 maint: DOI inactive as of April 2026 (link)