Hello! In our previous lesson, we established a profound connection between the two sides of the AdS/CFT correspondence: the symmetries match perfectly. We saw that the conformal group $SO(d,2)$ governing a $d$-dimensional CFT is precisely the isometry group of $(d+1)$-dimensional Anti-de Sitter space.
Today, we will build on this foundation to make the correspondence more concrete and computationally useful. We will move beyond symmetries to establish a direct "dictionary" that maps the physical objects in one theory to those in the other. This lesson will introduce the operator-field correspondence, a central pillar of holography. Our goal is to derive the precise mathematical relationship between the scaling dimension of an operator in the boundary CFT and the mass of its corresponding field in the AdS bulk.
1. The Fundamental Postulate: The GKPW Dictionary
The core prescription for how to translate between the two theories was formulated by Gubser, Klebanov, Polyakov, and by Witten, and is often called the GKPW dictionary. It provides an explicit equivalence between the partition functions of the two theories.
The central idea is that local, gauge-invariant operators in the boundary CFT, let's call them $O(x)$, are dual to dynamical fields in the bulk, let's call them $\phi(x,z)$. The boundary value of the bulk field, $\phi_0(x)$, acts as a source for the corresponding CFT operator.
Introduction to the AdS/CFT Duality - Martin Ammon
To begin, let's watch a segment from Dr. Martin Ammon's lecture, 'Introduction to the AdS/CFT Duality'. He provides a very clear, conceptual overview of this mapping and how the dynamics of the two theories are related through their generating functionals.
Please watch the segment from 20:44 to 24:37. Focus on how local gauge-invariant operators are mapped to bulk fields, and the statement of the correspondence in terms of the CFT generating functional and the string theory partition function.
As Dr. Ammon explained, the formal statement of the correspondence is an equality of partition functions:
On the left, we have the generating functional for correlation functions of the operator $O(x)$ in the CFT, sourced by $\phi_0(x)$. On the right, we have the full quantum gravity (string theory) partition function in the bulk, with the constraint that the field $\phi$ approaches the value $\phi_0$ at the boundary of AdS.
This statement becomes computationally powerful in the limit we discussed previously (large $N$, large 't Hooft coupling $\lambda$). In this limit, the quantum gravity path integral is dominated by its classical saddle-point. This means we can approximate the right-hand side by the exponential of the on-shell classical supergravity action:
This leads to the celebrated approximation:
where $W_{CFT}$ is the generator of connected correlation functions. This incredible result implies that we can compute quantum correlation functions in a strongly coupled QFT by solving classical field equations in a weakly curved higher-dimensional spacetime.
2. Deriving the Mass-Dimension Relation
Let's now use this framework to find the relationship between the mass of a bulk field and the dimension of its dual operator. We'll do this by studying the simplest case: a massive scalar field $\phi$ in $AdS_{d+1}$.
The procedure is as follows:
- Write down the action for the scalar field in AdS.
- Derive its classical equation of motion.
- Analyze the behavior of the solutions near the AdS boundary ($z \to 0$).
- Interpret the two independent solutions in terms of the source and the response (VEV) of the dual CFT operator.
This analysis will yield a direct relationship between the bulk mass $m$ and a parameter $\Delta$, which we will identify as the conformal dimension of the dual operator.
The following text from Christopher Herzog's lecture notes, 'AdS/CFT', walks through this derivation with mathematical clarity. It introduces the indicial equation, which is the key to the mass-dimension relation.
Please read the section 'How AdS/CFT works' from page 8 up to the paragraph before equation (39) on page 9. Focus on the action for the scalar field (eq. 35), the equation of motion (eq. 37), and the derivation of the indicial equation (eq. 38) that results from a power-law ansatz near the boundary.
As derived in the notes, the equation of motion for a massive scalar field in Euclidean $AdS_{d+1}$ (in Poincaré coordinates) is:
By analyzing the behavior near the boundary $z \to 0$ and assuming a solution of the form $\phi \sim z^\Delta$, we arrive at the indicial equation:
This is the central result of our lesson. This simple quadratic equation provides a precise dictionary entry, relating the mass-squared $m^2$ of a scalar field in the bulk to a parameter $\Delta$.
The equation has two solutions for $\Delta$:
The general asymptotic solution for $\phi$ near the boundary is a superposition of these two behaviors:
Here, we've used the fact that $d-\Delta_+ = \Delta_-$. The term with the smaller exponent, $z^{d-\Delta_+}$, dominates as $z \to 0$. This is the "non-normalizable" mode, and its coefficient $A(x)$ is identified with the source $\phi_0(x)$ for the dual operator. The sub-leading term, $z^{\Delta_+}$, is the "normalizable" mode, and its coefficient $B(x)$ is related to the vacuum expectation value $\langle O(x) \rangle$.
The parameter $\Delta = \Delta_+$ is identified as the scaling dimension of the dual operator $O(x)$. This can be understood from the AdS isometry $x^\mu \to \lambda x^\mu, z \to \lambda z$. Under this scaling, the field transforms as $\phi(x,z) \to \phi(\lambda x, \lambda z)$. For the source term $\phi_0(x)z^{d-\Delta}$ to be invariant, $\phi_0(x)$ must scale as $\lambda^{-\Delta}$, which is precisely the transformation property of an operator with dimension $\Delta$.
3. Physical Interpretation and Consequences
We've derived the fundamental relationship. Now let's explore its physical meaning and some immediate consequences.
AdS/CFT Correspondence, Part 1 - Juan Maldacena
Let's watch a clip from Juan Maldacena's seminal lecture, 'AdS/CFT Correspondence, Part 1'. He explains this same relationship from a slightly different, more physical perspective, relating the energy of a massive particle sitting in the center of AdS to the dimension of the corresponding state/operator on the boundary.
Please watch from 28:27 to 33:10. Pay close attention to how he introduces the state-operator mapping, connects the energy of a bulk particle to the conformal dimension, and gives the formula for a scalar field in d=4. Note his example of a massless particle corresponding to an operator of dimension 4.
The Dictionary Entry
Let's summarize our key finding for a scalar field:
A scalar field of mass $m$ in $AdS_{d+1}$ is dual to a scalar primary operator $O$ of conformal dimension $\Delta$ in $CFT_d$, where the two are related by:
Unitarity and the Breitenlohner-Freedman (BF) Bound
A remarkable consistency check arises when we consider stability and unitarity.
- In the Bulk (AdS): A scalar field in AdS can have a negative mass-squared without being a tachyon (i.e., without causing an instability). The curvature of AdS provides a stabilizing potential. The condition for stability is known as the Breitenlohner-Freedman (BF) bound: $m^2L^2 \ge -d^2/4$.
- On the Boundary (CFT): In a unitary CFT, the dimensions of scalar operators are bounded from below. For a $d$-dimensional CFT, this unitarity bound is $\Delta \ge (d-2)/2$.
Let's see if these two bounds are consistent via our dictionary formula. The BF bound $m^2L^2 = -d^2/4$ translates to:
This is not quite the unitarity bound. However, looking at the solution $\Delta = \frac{d}{2} + \sqrt{(\frac{d}{2})^2 + m^2L^2}$, we see that the minimum possible value of $\Delta$ (when $m^2L^2 > -d^2/4$) is just above $d/2$. The bound $\Delta \ge (d-2)/2$ is weaker and always satisfied in this range. The saturation of the unitarity bound, $\Delta=(d-2)/2$, corresponds to a free scalar field in the CFT. Using our formula, this dimension corresponds to a bulk mass of $m^2L^2 = \frac{d-2}{2}(\frac{d-2}{2}-d) = -(\frac{d+2}{2})(\frac{d-2}{2})$.
For a more comprehensive view, including fields with spin, the following resource is excellent.
Introduction to the AdS/CFT correspondence
These lecture notes, 'Introduction to the AdS/CFT correspondence', provide a concise summary of the mass-dimension relation for fields of various spins and an excellent discussion of the BF bound and an interesting subtlety called 'alternate quantization'.
Please read Section 2.2 'Physics in the bulk' (pages 24-29). Pay special attention to the formula in equation (2.34) and the discussion points that follow, including the table of mass-dimension relations for fields of different spins.
The table in the notes is particularly valuable. It shows, for example, that a massless spin-1 field (a gauge boson) in the bulk corresponds to a conserved current with dimension $\Delta = d-1$ in the CFT, and a massless spin-2 field (the graviton) corresponds to the stress-energy tensor with dimension $\Delta = d$. These are exactly the dimensions required by conformal symmetry for these special operators.
4. A Powerful Check: Protected Operators in $\mathcal{N}=4$ SYM
The most striking checks of the AdS/CFT correspondence come from comparing quantities that are "protected" by supersymmetry and thus do not change as the coupling is varied. This allows for a direct comparison between a weak-coupling calculation in the CFT and a strong-coupling (classical gravity) calculation in the bulk.
In $\mathcal{N}=4$ SYM, there is a class of operators known as BPS operators whose dimensions are fixed by their R-charge due to the constraints of the superconformal algebra. A primary example is the family of chiral primary operators:
where $\phi^I$ are the six scalar fields of the theory, and the indices are symmetrized and made traceless. Supersymmetry protects the dimension of this operator to be exactly $\Delta = p$, for any value of the 't Hooft coupling.
On the gravity side, these operators are dual to Kaluza-Klein modes of the supergravity fields on the $S^5$. By performing a KK reduction, one can calculate the masses of these bulk fields. For the fields dual to $O_p$, the result is:
Let's check this against our dictionary for $d=4$:
The results match perfectly! This is a highly non-trivial, quantitative success of the correspondence.
Test your understanding!
Consider the operator $O_2 = \text{Tr}(\phi^I \phi^J)$ (with indices symmetrized and traceless) in 4D $\mathcal{N}=4$ SYM.
- What is its protected conformal dimension $\Delta$?
- Using the dictionary, what is the required mass-squared ($m^2L^2$) of its dual scalar field in $AdS_5$?
Show answer
- The operator $O_2$ has $p=2$ scalar fields, so its protected dimension is $\Delta = 2$.
- Using the mass-dimension relation for $d=4$, $m^2L^2 = \Delta(\Delta-4)$, we get $m^2L^2 = 2(2-4) = -4$. This matches the Breitenlohner-Freedman bound in $d=4$ ($m^2L^2 \ge -4$) and corresponds to the lightest massive Kaluza-Klein scalar mode on the $S^5$.
For a deeper dive into why these operator dimensions are protected, the following section from Herzog's notes is invaluable. It uses the structure of the superconformal algebra to argue that the dimensions of BPS operators are fixed. This is a rigorous argument that you might find particularly insightful.
Please read Section 4, 'A Basic Check of AdS/CFT', and Section 4.1 'The Long Version' (pages 18-23). The 'short version' in the first few paragraphs gives the main idea. The 'long version' provides the detailed algebraic argument, focusing on how the anticommutation relation {Q, S} links the dimension D to the R-charge R, thereby protecting it.
Conclusion
In this lesson, we have forged the first and most crucial link in the holographic dictionary: the operator-field correspondence. We've gone from a general statement about matching symmetries to a precise, quantitative relationship between the properties of objects living in different dimensions.
Key Takeaways:
- The GKPW dictionary states that the CFT generating functional for an operator $O$ is equivalent to the bulk gravity partition function for the dual field $\phi$, where the boundary value of $\phi$ acts as the source for $O$.
- In the classical gravity limit, this allows us to compute quantum CFT correlators from the on-shell classical action in AdS.
- By analyzing the asymptotic solutions to a bulk scalar field's equation of motion, we derived the fundamental mass-dimension relation: $m^2L^2 = \Delta(\Delta-d)$.
- This relation beautifully maps the stability bound for bulk fields (the BF bound) to the unitarity bound for CFT operators.
- The correspondence passes a powerful quantitative test by correctly matching the dimensions of protected BPS operators in $\mathcal{N}=4$ SYM with the masses of their dual Kaluza-Klein modes in the bulk.
Next Lesson Preview:
Now that we have this dictionary entry, we can begin to use it for calculations. How do we compute something as fundamental as the two-point function $\langle O(x)O(y) \rangle$ from the bulk? In the next lesson, we will see how the on-shell action for a bulk field, determined by its bulk-to-boundary propagator, gives precisely the correct power-law form for the two-point function of the dual CFT operator.