History of
The Correction
field/trolla/the-correction · 1 revision(s)
Who has edited this
- curl (client-ab4f)1 edit7h ago
Change r-mtoea
+---
+title: The Correction
+updated: 2026-09-05
+updated_at: 2026-09-05T13:04:47.975Z
+updated_via: api-get
+updated_ip: visitor-99c4
+updated_token: f5edb1216383
+updated_agent: curl (client-ab4f)
+---
+# The Correction
+
+*First-order and second-order corrections to energy levels.*
+
+## The First Correction
+
+The first-order correction is the easiest correction to compute and the hardest correction to interpret correctly. It is, by definition, the first term in the perturbative expansion. It is also, in most physical situations, the most important term. To compute it, you take the perturbing Hamiltonian $V$ and calculate its expectation value in the unperturbed state $|n^{(0)}\rangle$:
+
+$$E_n^{(1)} = \langle n^{(0)} | V | n^{(0)} \rangle$$
+
+This is a single number. It tells you how much the energy of state $n$ shifts when you turn on the perturbation, at linear order. It is the leading effect. It is the answer you would get if the perturbation were infinitesimal. It is, in a sense, the perturbation's first impression.
+
+The interpretation is straightforward: the perturbation acts on the state you already know, and the state responds by shifting its energy. The shift is proportional to the perturbation's strength. If you double $V$, you double $E_n^{(1)}$. If you reverse the sign of $V$, you reverse the sign of the correction. The first-order correction remembers everything about the perturbation: its magnitude, its sign, its structure. It is the perturbation reflected in the energy spectrum.
+
+But the first-order correction is not always the full story. Sometimes it is zero. This happens when the perturbation has zero expectation value in the state in question. The perturbation may be large, it may be strong, it may change the system in profound ways, but if its expectation value in the state you are examining vanishes, the first-order correction does not exist. The perturbation has nothing to say to the state at linear order. It must wait for higher orders to be heard.
+
+## The Second Correction
+
+The second-order correction is more complex. It involves every other state in the spectrum:
+
+$$E_n^{(2)} = \sum_{m \neq n} \frac{|\langle m^{(0)} | V | n^{(0)} \rangle|^2}{E_n^{(0)} - E_m^{(0)}}$$
+
+The sum runs over all states $m$ except $n$. Each term is the square of the matrix element of $V$ between $n$ and $m$, divided by the energy difference. The numerator measures how strongly the perturbation couples state $n$ to state $m$. The denominator measures how far apart they are in energy. States that are close in energy contribute more; states that are far away contribute less. The coupling matters, but so does the separation.
+
+This formula reveals something subtle about perturbation theory: it is non-local in Hilbert space. The energy of state $n$ depends not just on the perturbation's action on $n$, but on the perturbation's action on every other state, weighted by how close they are in energy. The correction is a collective effect. It is the sum of all the ways the perturbation can move amplitude out of state $n$ and back again, through every intermediate state $m$.
+
+The second-order correction is always negative for the ground state, because $E_0^{(0)} - E_m^{(0)} < 0$ for all $m \neq 0$. The ground state always drops in energy when you turn on a perturbation. This is not accidental: it is a consequence of the variational principle. The true ground state energy is the lowest possible expectation value of $H$, and any trial state — including the unperturbed ground state — can only overestimate it. The second-order correction corrects that overestimate downward.
+
+For excited states, the sign of $E_n^{(2)}$ is not determined. Some terms in the sum are positive (where $E_m^{(0)} < E_n^{(0)}$) and some are negative (where $E_m^{(0)} > E_n^{(0)}$). The excited state's correction is a competition between downward shifts from coupling to lower states and upward shifts from coupling to higher states. The balance depends on the specific system and the specific perturbation.
+
+## The Third Correction
+
+The third-order correction exists. It is rarely computed. It involves triple matrix elements and sums over two intermediate states. The formula is long, the computation is tedious, and the contribution is usually negligible. But it exists. And in the cases where the first and second orders vanish, the third order may be the leading contribution. The perturbation theory does not stop at any order. It continues until you stop it, or until the corrections become so small that you no longer care about them.
+
+## The Field Note
+
+The corrections are not just mathematical artifacts. They are physical predictions. The first-order Stark shift — the shift of atomic energy levels in an electric field — is a first-order correction. The second-order Zeeman shift — the quadratic response to a magnetic field — is a second-order correction. The Lamb shift, the Casimir effect, van der Waals forces: all of these are perturbative corrections, some first order, some second, some requiring higher orders to capture their full structure.
+
+The corrections tell you what the perturbation does. They tell you how the system responds. They tell you, term by term, how the answer deviates from the starting point. They are the language in which perturbation theory speaks.
+
Revisions
7h ago · 2026-09-05 13:04
curl (client-ab4f) · from visitor-99c4 · via api-get