What Is a Tensor?
In mathematics and physics a tensor is a geometric object that generalizes scalars (zerorank tensors) and vectors (firstrank tensors) to higher dimensions. Tensors are defined by how they transform under a change of coordinates. This transformation property guarantees that the physical meaning of a tensor does not depend on the particular coordinate system used to describe it.
The rank (or order) of a tensor tells how many indices it carries. A rank2 tensor, for instance, can be written as T^{ij}, where each index can take the values 1,,n for an ndimensional space. When the indices are placed one up and one down (e.g., T^{i}{}_j) the object is a mixedtype tensor.
Basic Operations in Tensor Algebra
- Addition and subtraction Tensors of the same rank and type can be added componentwise.
- Scalar multiplication Multiplying a tensor by a scalar scales every component.
- Tensor product The outer product of a rankp tensor A^{i_1i_p} with a rankq tensor B^{j_1j_q} creates a rank(p+q) tensor C^{i_1i_p j_1j_q}=A^{i_1i_p}B^{j_1j_q}.
- Contraction Summing over a pair of one upper and one lower index reduces the rank by two. Example: C^{i}=A^{ij}B_{j} is the contraction of the product A^{ij}B_{j}.
- Symmetrization and antisymmetrization By adding (or subtracting) permutations of indices one can obtain symmetric (e.g., S^{(ij)}=(T^{ij}+T^{ji})/2) or antisymmetric tensors (e.g., A^{[ij]}=(T^{ij}-T^{ji})/2).
Metric Tensor and Raising/Lowering Indices
The metric tensor g_{ij} encodes the inner product of the space. In Euclidean space it reduces to the Kronecker delta \delta_{ij}, while in curved manifolds (e.g., general relativity) it carries information about curvature. Using the metric one can raise or lower indices: V_i = g_{ij}V^j, \qquad V^i = g^{ij}V_j. This operation is essential for converting covariant to contravariant components and viceversa.
From Algebra to Calculus: Tensor Derivatives
Tensor calculus extends the algebraic operations to differentiation. The central object is the covariant derivative, denoted \nabla_k or \;{;k}. For a vector field V^i the covariant derivative is \nabla_k V^i = \partial_k V^i + \Gamma^{i}_{\;kj}V^j, where \Gamma^{i}_{\;kj} are the Christoffel symbols of the connection. The additional term adjusts for the change of the basis vectors when moving from point to point.
The covariant derivative of a general tensor follows a similar pattern: each upper index receives a + term, each lower index a term. For example, for a rank(1,1) tensor T^{i}{}_{j}: \nabla_k T^{i}{}_{j} = \partial_k T^{i}{}_{j} + \Gamma^{i}_{\;km}T^{m}{}_{j} - \Gamma^{m}_{\;kj}T^{i}{}_{m}.
Important Tensor Identities
- Metric compatibility: \nabla_k g_{ij}=0. This property ensures that lengths and angles are preserved under parallel transport.
- Riemann curvature tensor: R^{i}{}_{jkl} =\partial_k\Gamma^{i}_{\;jl} -\partial_l\Gamma^{i}_{\;jk} +\Gamma^{i}_{\;km}\Gamma^{m}_{\;jl} -\Gamma^{i}_{\;lm}\Gamma^{m}_{\;jk}. It measures the noncommutativity of covariant derivatives.
- Ricci tensor: contraction of the Riemann tensor, R_{jl}=R^{i}{}_{jil}.
- Scalar curvature: trace of the Ricci tensor, R=g^{jl}R_{jl}.
Applications in Physics and Engineering
General Relativity: Einsteins field equations G_{ij}=8\pi T_{ij} relate the Einstein tensor G_{ij}=R_{ij}-\frac12 g_{ij}R (built from curvature tensors) to the stressenergy tensor T_{ij}. These equations are fundamentally tensorial, guaranteeing invariance under any coordinate transformation.
Continuum Mechanics: Stress and strain are represented by secondorder tensors \sigma_{ij} and \varepsilon_{ij}. Hookes law for an anisotropic material reads \sigma_{ij}=C_{ijkl}\,\varepsilon_{kl}, where C_{ijkl} is the fourthorder elasticity tensor.
Electromagnetism: The electromagnetic field tensor F_{\mu\nu} encapsulates electric and magnetic fields. Maxwells equations acquire a compact form \nabla_{[\lambda}F_{\mu\nu]}=0, \qquad \nabla^\mu F_{\mu\nu}= \mu_0 J_{\nu}.
Computer Graphics: Transformations of points and normals are carried out with matrices (rank2 tensors). Homogeneous coordinates, curvature calculations for surfaces, and anisotropic shading all rely on tensor concepts.
Why Tensor Calculus Matters
Tensor calculus provides a language that is independent of the particular coordinate grid used to describe a problem. This coordinatefree approach simplifies the derivation of physical laws, ensures that predictions are objectively meaningful, and makes it easier to transition between Cartesian, curvilinear, and even abstract manifolds.
Moreover, many modern fieldssuch as machine learning (tensorbased data structures), quantum field theory (operatorvalued tensors), and differential geometryrely on the same foundational ideas. Mastery of tensor algebra and calculus therefore opens the door to a broad range of advanced scientific and engineering topics.
Further Reading
- W.M.Kelley, Tensor Calculus, Dover Publications, 2003.
- J.J.Sakurai, Modern Quantum Mechanics, especially the sections on tensor operators.
- R.M.Carter, General Relativity and Gravitation, Cambridge University Press.
- Online resource: Wikipedia Tensor.
