1:
2: \chapter{Hyperbolic space} \label{ch4}
3:
4: In this chapter, which is based on \cite{Ambjorn:2006hu}, we go back to an analysis of pure CDT.
5: As in most approaches to quantum gravity, the CDT method was originally developed
6: for the quantization of compact manifolds. In the following however we generalize the boundary conditions
7: of two dimensional CDT so that the typical geometries in the path integral have an infinite volume.
8: Particularly, it is shown that given such boundary conditions a classical geometry with constant negative curvature
9: and superimposed quantum fluctuations emerges from the background independent path integral. Furthermore,
10: one can choose the boundary conditions such that the relative fluctuations become small in a concrete manner.
11: To the knowledge of the author this is one of the few cases where a semiclassical geometry emerges from a genuinely
12: background setup that can at the same time be studied by analytical methods. Another example were a semiclassical
13: background emerges dynamically from a background independent scheme is four dimensional quantum gravity defined
14: by causal dynamical triangulations. This model is too complicated to study with analytical methods, nevertheless
15: it is one of the most promising attempts to formulate a realistic theory of quantum gravity
16: (for a recent account see \cite{Ambjorn:2006jf}).
17:
18:
19: \section{Non compact manifolds}\label{sec:non compact manifolds}
20:
21: %Two-dimensional quantum gravity is not much of a gravity theory in
22: %the sense that there are no propagating gravitons. It has
23: %nevertheless been a fertile playground when it comes to testing
24: %various aspects of diffeomorphism-invariant theories, and it is
25: %potentially important for string theory which can be viewed as
26: %two-dimensional quantum gravity coupled to specific, conformal
27: %invariant matter fields.
28:
29: As mentioned before, 2d quantum gravity is intimately related with the study of non-critical string theories.
30: The studies where the quantum gravity aspect has been emphasized mostly consider two dimensional Euclidean quantum
31: gravity with compact spacetime. The study of 2d Euclidean quantum
32: gravity with non-compact spacetime was initiated by the
33: Zamolodchikovs (ZZ) \cite{Zamolodchikov:2001ah} when they showed how to use
34: conformal bootstrap and the cluster-decomposition properties to
35: quantize Liouville theory on the pseudo-sphere (the Poincar\'{e} disc).
36:
37: Martinec \cite{Martinec:2003ka} and Seiberg et al.\ \cite{Seiberg:2003nm} showed how
38: the work of ZZ fitted into the framework of non-critical string
39: theory, where the ZZ-theory could be reinterpreted as special
40: branes, now called ZZ-branes. Let $W_\tL(\tX) $ be the ordinary
41: disc amplitude for $2d$ Euclidean gravity on a compact
42: spacetime. $\tX$ denotes the boundary cosmological constant of
43: the disc and $\tL$ the cosmological constant. It was found that
44: the ZZ-brane of 2d Euclidean gravity was associated with the zero
45: of
46:
47: \beq\label{0.0}
48: W_\tL(\tX) = (\tX -\oh \sqrt{\tL}) \sqrt{\tX + \sqrt{\tL}}.
49: \eeq
50:
51: At first sight this is somewhat surprising since
52: from a world-sheet point of view the disc is compact while the
53: Poincar\'{e} disc is non-compact. In \cite{Ambjorn:2004my} and \cite{Ambjorn:2007iq} it was shown
54: how it could be understood in terms of world sheet geometry, i.e.~from a
55: 2d quantum gravity point of view. When the boundary cosmological
56: constant $\tX$ reaches the value $\tX \equ \sqrt{\tL}/2$ where the
57: disc amplitude $W_\tL(\tX) \equ 0$, the geodesic distance from a
58: generic point on the disc to the boundary diverges, in this way
59: effectively creating a non-compact spacetime.
60:
61: Here we show that the same phenomenon occurs in two dimensional
62: quantum gravity from causal dynamical triangulations.
63:
64: \section{The hyperbolic plane from CDT}
65:
66: Recall that the propagator with one mark on the initial boundary is given by
67: \beq\label{2.a3}
68: G_\La (X,Y;T) = \frac{\bX^2(T,X) -\La}{X^2-\La} \; \frac{1}{\bX(T,X) +Y},
69: \eeq
70: where $\bX(T,X) $ is the solution of the characteristic equation
71: \beq\label{2.3}
72: \frac{\d \bX}{\d T} = -(\bX^2-\La),~~~\bX(0,X) =X,
73: \eeq
74: giving
75: \beq\label{2.4}
76: \bX(t,X) = \SL \coth \SL(t+t_0),~~~~X=\SL \coth \SL \,t_0.
77: \eeq
78: Note that although different in appearance this expression is equivalent to \rf{eq:xbar}.
79: Viewing $G_\La(X,Y;T) $ as a propagator with X and Y as coupling constants,
80: $\bX(T) $ can be viewed as a ``running'' boundary cosmological
81: constant, $T$ being the scale. If $X > - \SL$ then
82: $\bX(T) \to \SL$ for $T \to \infty$, $\SL$ being a ``fixed point''
83: (a zero of the ``$\b$-function'' $-(\bX^2-\La) $ in eq.\ \rf{2.3}).
84:
85: Let $L_1$ denote the length of the entry boundary and $L_2$ the
86: length of the exit boundary. Rather than consider a situation
87: where the boundary cosmological constant $X$ is fixed we can
88: consider $L_1$ as fixed. We denote the corresponding propagator
89: $G_\La (L_1,Y;T) $. Similarly we can define $G_\La(X,L_2;T) $
90: and $G_\La(L_1,L_2;T) $. They are related by Laplace transformations.
91: For instance:
92: \beq\label{2.a5}
93: G_\La(X,Y;T) = \int_0^\infty \d L_2 \int_0^\infty \d L_1\;
94: G(L_1,L_2;T) \;\e^{-XL_1-YL_2},
95: \eeq
96: and one has the following composition rule for the propagator:
97: \beq\label{2.a6}
98: G_\La (X,Y;T_1+T_2) = \int_0^\infty \d L \;
99: G_\La (X,L;T_1) \,G(L,Y,T_2).
100: \eeq
101:
102: We can now calculate the expectation value of the length of the
103: spatial slice at proper time $t \in [0,T]$:
104:
105: \beq\label{2.a7}
106: \la L(t) \ra_{X,Y,T} = \frac{1}{G_\La (X,Y;T) } \int_0^\infty \d L\;G_\La (X,L;t) \;L\; G_\La (L,Y;T-t).
107: \eeq
108:
109: In general there is no
110: reason to expect $\la L(t) \ra$ to have a classical limit.
111: Consider for instance the situation where $X$ and $Y$ are larger
112: than $\SL$ and where $T \gg 1/\SL$. The average boundary lengths
113: will be of order $1/X$ and $1/Y$. But for $ 0 \ll t \ll T $ the
114: system has forgotten everything about the boundaries and the
115: expectation value of $L(t) $ is, up to corrections of order
116: $e^{-2\SL t}$ or $e^{-2\SL (T-t) }$, determined by the ground state
117: of the effective Hamiltonian $H_{ef\!f}$ corresponding to
118: $G_\La(X,Y;T) $. One finds for this ground
119: state $\la L \ra = 1/\SL$. This picture is confirmed by an
120: explicit calculation using eq.\ \rf{2.a7} as long as $X,Y > \SL$.
121: The system is thus, except for boundary effects, entirely
122: determined by the quantum fluctuations of the ground state of
123: $H_{ef\!f}$.
124:
125: We will here be interested in a different and more interesting
126: situation where a non-compact spacetime is obtained as a limit of
127: the compact spacetime described by \rf{2.a7}. Thus we want to
128: take $T \to \infty$ and at the same time also the length of the
129: boundary corresponding to proper time $T$ to infinity. Since $T
130: \to \infty$ forces $\bX(T,X) \to \SL$ it follows from \rf{2.a3}
131: that the only choice of boundary cosmological constant $Y$
132: independent of $T$, where the length $\la L(T) \ra_{X,Y,T}$ goes
133: to infinity for $T \to \infty$ is $Y\equ \mi \SL$ (fig.~ \ref{fig:criticalboundary}), since we have:
134:
135: %
136: \begin{figure}[t]
137: \begin{center}
138: \includegraphics[width=4in]{criticalboundary}
139: \caption{For $Y = Y_c =-\sqrt{\La}$ the length of the final boundary diverges as $T \rightarrow \infty$.}
140: \label{fig:criticalboundary}
141: \end{center}
142: \end{figure}
143: %
144:
145: \beq\label{2.a8}
146: \la L(T) \ra_{X,Y,T} = -\frac{1}{G_\La (X,Y;T) } \, \frac{\prt G_\La (X,Y;T) }{\prt Y} = \frac{1}{\bX(T,X) +Y}.
147: \eeq
148:
149: With the choice $Y \equ - \SL$ one obtains from \rf{2.a7} in the
150: limit $T \to \infty$:
151:
152: \beq\label{2.a9}
153: \la L(t) \ra_{X} = \frac{1}{\SL} \; \sinh \left(2\SL(t+t_0(X) ) \right),
154: \eeq
155:
156: where $t_0(X) $ is defined in eq.\ \rf{2.4}.
157:
158: We have called $L_2$ the (spatial) length of the boundary
159: corresponding to $T$ and $\la L(t) \ra_X$ the spatial length of a
160: time-slice at time $t$ in order to be in accordance with earlier
161: notation \cite{Nakayama:1993we,Ambjorn:1998xu}, but starting from a lattice
162: regularization and taking the continuum limit $L$ is only
163: determined up to a constant of proportionality which we fix by
164: comparing with a continuum effective action. In section \rf{subsec:Hamiltonians in causal quantum gravity}
165: we showed that such a comparison leads to the identification
166: of $L$ as $L_{cont}/\pi$ and we are led to the following
167:
168: \beq\label{2.a10}
169: L_{cont}(t) \equiv \pi \la L(t) \ra_X = \frac{\pi}{\SL} \; \sinh \left(2\SL(t+t_0(X) )\right).
170: \eeq
171:
172: Consider the
173: classical surface where the intrinsic geometry is defined by
174: proper time $t$ and spatial length $L_{cont}(t) $ of the curve
175: corresponding to constant $t$. It has the line element
176:
177: \beq\label{2.a11}
178: \d s^2 = \d t^2 + \frac{L_{cont}^2}{4\pi^2}\; \d \th^2 = \d t^2 + \frac{\sinh^2 \left(2\SL (t+t_0(X) ) \right) }{4 \La} \;\d \th^2,
179: \eeq
180:
181: where $t \ge 0$ and $t_0(X) $ is a function of the
182: boundary cosmological constant $X$ at the boundary corresponding
183: to $t \equ 0$ (see eq.\ \rf{2.4}). What is remarkable about
184: formula \rf{2.a11} is that the surfaces for different boundary
185: cosmological constants $X$ can be viewed as part of the same
186: surface, the Poincar\'{e} disc with curvature $R= -8\La$, since $t$
187: can be continued to $t= -t_0 $. The Poincar\'{e} disc itself is
188: formally obtained in the limit $X \to \infty$ since an infinite
189: boundary cosmological constant will contract the boundary to a
190: point.
191:
192: \section{The classical effective action}
193:
194: In this section we make a small digression to the ``classical'' theory
195: and show that the emergence of the hyperbolic plane is natural from this point of view.
196: In section \rf{subsec:Hamiltonians in causal quantum gravity} we discussed the derivation
197: of the quantum Hamiltonian of causal quantum gravity from the following classical action.
198: %
199: \beq\label{3.4}
200: S_\k = \int_0^T \d t \left(\frac{\dot{l}^2(t) }{4l(t) } +
201: \La l(t) + \frac{\k}{l}\right).
202: \eeq
203: %
204: To make contact with the inherently quantum calculation by causal dynamical triangulations
205: it is interesting to look at the classical behavior corresponding to this action.
206: The classical solutions corresponding to action \rf{3.4} are
207: %
208: \bea
209: l(t) &=& \frac{\sqrt{\k}}{\SL} \; \sinh 2\SL t,~~~~~~~~~~\k>0~~ \mbox{elliptic case}, \label{3.6a}\\
210: l(t) &=& \frac{\sqrt{-\k}}{\SL} \; \cosh 2\SL t,~~~~~~~\k<0~~\mbox{hyperbolic case}, \label{3.6b}\\
211: l(t) &=& \e^{2\SL t},~~~~~~~~~~~~~~~~~~~~~~~\k=0 ~~\mbox{parabolic case}, \label{3.6c}
212: \eea
213: %
214: all corresponding to cylinders with constant negative curvature $-8 \La$.
215: In the elliptic case, where $t$ must be larger than zero, there is
216: a conical singularity at $t =0$ unless $\k = 1$. For $\k \equ 1$
217: the geometry is regular at $t=0$ and this value of $\k$
218: corresponds precisely to the Poincar\'{e} disc, $t=0$ being the
219: ``center'' of the disc. So we see that for $\k \equ 1$ the classical solution
220: coincides nicely with the emergent geometry derived in the previous section
221:
222: \section{Quantum fluctuations}
223:
224: In many ways it is more natural to fix the boundary cosmological
225: constant than to fix the length of the boundary. However, one pays
226: the price that the fluctuations of the boundary size are large, in
227: fact of the order of the average length of the boundary itself
228: \footnote{This is true also in Liouville quantum theory, the
229: derivation is essentially the same as that given in \rf{5.1}, as is
230: clear from \cite{Ambjorn:2004my}.}: from \rf{2.a8} we have
231:
232: \beq\label{5.1}
233: \la L^2(T) \ra_{X,Y;T} - \la L(T) \ra^2_{X,Y;T} = -\frac{\prt \la L(T) \ra_{X,Y;T}}{\prt Y} = \la L(T) \ra^2_{X,Y;T}.
234: \eeq
235:
236: Such large fluctuations are also present around $\la L(t) \ra_{X,Y;T}$
237: for $t< T$. From this point of view it is even more remarkable that
238: $\la L(t) \ra_{X,Y=- \SL;T=\infty}$ has such a nice semiclassical
239: interpretation. Let us now by hand fix the boundary lengths $L_1$
240: and $L_2$. This is done in the Hartle-Hawking Euclidean path
241: integral when the geometries $[g]$ are fixed at the boundaries
242: \cite{Hartle:1983ai}. For our one-dimensional boundaries the geometries at
243: the boundaries are uniquely fixed by specifying the lengths of the
244: boundaries, and the relation between the propagator with fixed
245: boundary cosmological constants and with fixed boundary lengths is
246: given by a Laplace transformation as shown in eq.\ \rf{2.a5}. Let
247: us for simplicity analyze the situation where we take the length
248: $L_1$ of the entrance loop to zero by taking the boundary
249: cosmological constant $X \to \infty$. Using the decomposition
250: property \rf{2.a6} one can calculate the connected ``loop-loop''
251: correlator for fixed $L_2$ and $0< t \leq t+\Del < T$,
252:
253: \beq\label{5.a1}
254: \la L(t) L(t+\Del) \ra^{(c) }_{L_2,T} \equiv \la L(t+\Del) L(t) \ra_{L_2,T}-\la L(t) \ra \la L(t+\Del) \ra_{L_2,T}.
255: \eeq
256:
257: One finds
258:
259: \bea\label{5.a2}
260: \la L(t) L(t\pl\Del) \ra^{(c) }_{L_2,T} &=& \frac{2}{\La} \frac{\sinh^2 \SL t \sinh^2 \SL (T\mi (t\pl\Del) ) }{\sinh^2 \SL T}+
261: \\
262: && \frac{2L_2}{\SL} \frac{\sinh^2 \SL t \sinh\SL (t\pl\Del) \sinh\SL(T \mi(t\pl\Del) ) }{\sinh^3 \SL T}.\no
263: \eea
264:
265: We also note that
266:
267: \beq\label{5.b2}
268: \la L(t) \ra_{L_2,T}= \frac{2}{\SL} \frac{\sinh \SL t \sinh \SL (T\mi t) }{\sinh \SL T}+ L_2\frac{\sinh^2 \SL t}{\sinh^2 \SL T}.
269: \eeq
270:
271: For fixed $L_2$ and $T \to \infty$ we obtain
272:
273: \beq\label{5.a3} \la
274: L(t) L(t+\Del) \ra^{(c) }_{L_2} = \frac{1}{2\La} \; \e^{-2\SL \Del } \left(1-\e^{-2\SL t} \right) ^2 \eeq and \beq\label{5.b3} \la L(t) \ra_{L_2}=\frac{1}{\SL}\left( 1-\e^{-2\SL t}\right).
275: \eeq
276:
277: Eqs.\ \rf{5.a3} and \rf{5.b3} tell us that except for small $t$
278: we have $\la L(t) \ra_{L_2}\equ 1/\SL$. The quantum fluctuations
279: $\Del L(t) $ of $L(t) $ are defined by $(\Del L(t) ) ^2 = \la
280: L(t) L(t) \ra^{(c) }$. Thus the spatial extension of the universe is
281: just quantum size (i.e.~ $1/\SL$, $\La$ being the only coupling
282: constant) with fluctuations $\Del L(t) $ of the same size. The time
283: correlation between $L(t) $ and $L(t+\Del) $ is also dictated by the
284: scale $1/\SL$, telling us that the correlation between spatial
285: elements of size $1/\SL$, separated in time by $\Del$ falls of
286: exponentially as $e^{-2\SL \Del}$ . The above picture is precisely
287: what one would expect from the classical action, which is proportional
288: to the area and the boundary cosmological constants only, if we force $T$ to
289: be large and choose a $Y$ such that $\la L_2(T) \ra$ is not large,
290: the universe will be a thin tube, ``classically'' of zero width,
291: but due to quantum fluctuations of average width $1/\SL$.
292:
293: A more interesting situation is obtained if we choose $Y = -\SL$,
294: the special value needed to obtain a non-compact geometry in the
295: limit $T\to \infty$. To implement this in a setting where $L_2$ is
296: not allowed to fluctuate we fix $L_2(T) $ to the average value
297: \rf{2.a8} for $Y\equ \mi \SL$:
298:
299: \beq\label{5.2}
300: L_2(T) = \la L(T) \ra_{X,Y= -\SL;T} = \frac{1}{\SL} \; \frac{1}{\coth \SL T -1}.
301: \eeq
302:
303: From \rf{5.a2} and \rf{5.b2} we have in the limit $T \to \infty$:
304:
305: \beq\label{5.3}
306: \la L(t) \ra = \frac{1}{\SL} \; \sinh 2\SL t,
307: \eeq
308:
309: in accordance with \rf{2.a9}, and for the
310: ``loop-loop''-correlator
311:
312: \beq\label{5.4}
313: \la L(t+\Del) L(t) \ra^{(c) }= \frac{2}{\La}\;\sinh^2 \SL t= \frac{1}{\SL} \left( \la L(t) \ra -\frac{1}{\SL} \left(1-\e^{-2\SL t}\right) \right).
314: \eeq
315:
316: It is seen that the ``loop-loop''-correlator is independent of
317: $\Del$. In particular we have for $\Del \equ 0$:
318:
319: \beq\label{5.5}
320: (\Del L(t) ) ^2 \equiv \la L^2(t) \ra -\la L(t) \ra^2 \sim \frac{1}{\SL} \la L(t) \ra
321: \eeq
322:
323: for $t \gg 1/\SL$. The
324: interpretation of eq.\ \rf{5.5} is in accordance with the picture
325: presented below \rf{5.b3}: We can view the curve of length $L(t) $
326: as consisting of $N(t) \approx \SL L(t) \approx e^{2\SL t} $
327: independently fluctuating parts of size $1/\SL$ and each with a
328: fluctuation of size $1/\SL$. Thus the total fluctuation $\Del
329: L(t) $ of $L(t) $ will be of order $1/\SL \times \sqrt{N(t) }$,
330:
331: \beq\label{5.a5}
332: \frac{\Del L(t) }{\la L(t) \ra} \sim \frac{1}{\sqrt{\SL \la L(t) \ra}} \sim \e^{-\SL t},
333: \eeq
334:
335: i.e.\ the
336: fluctuation of $L(t) $ around $\la L(t) \ra$ is small for $t \gg
337: 1/\SL$. In the same way the independence of the
338: ``loop-loop''-correlator of $\Del$ can be understood as the
339: combined result of $L(t+\Del) $ growing exponentially in length
340: with a factor $e^{2\SL \Del}$ compared to $L(t) $ and, according to
341: \rf{5.a3}, the correlation of ``line-elements'' of $L(t) $ and
342: $L(t+\Del) $ decreasing by a factor $e^{-2\SL \Del}$.
343:
344: \section{Summary}
345:
346: We have described how the CDT quantization of 2d gravity for a
347: special value of the boundary cosmological constant leads to a
348: non-compact (Euclidean) AdS-like spacetime of constant negative
349: curvature dressed with quantum fluctuations. It is possible to
350: achieve this non-compact geometry as a limit of a compact geometry
351: as described above. In particular the assignment \rf{5.2} leads to
352: a simple picture where the fluctuation of $L(t) $ is small compared
353: to the average value of $L(t) $. In fact the geometry can be viewed
354: as that of the Poincar\'{e} disc with fluctuations correlated only
355: over a distance $1/\SL$.
356:
357: Our construction is similar to the analysis of $ZZ$-branes
358: appearing as a limit of compact 2d geometries in Liouville quantum
359: gravity \cite{Ambjorn:2004my}. In the CDT case the non-compactness came when
360: the running boundary cosmological constant $\bX(T) $ went to the
361: fixed point $\SL$ for $T \to \infty$. In the case of Liouville
362: gravity, represented by DT (or equivalently matrix models), the
363: non-compactness arose when the running (Liouville) boundary
364: cosmological constant ${\bX_{Liouville}(T) }$ went to the value
365: where the disc amplitude $W_\tL(\tX) \equ 0$, i.e.\ to $\tX \equ
366: \sqrt{\tL}/2$ (see eq.\ \rf{0.0}). It is the same process in the
367: two cases, since the relation between Liouville gravity and CDT is
368: well established and summarized by the mapping \cite{Ambjorn:1999fp}:
369:
370: \beq\label{6.1}
371: \frac{X}{\SL} = \sqrt{\frac{2}{3}}\; \sqrt{1+\frac{\tX}{\sqrt{\tL}}},
372: \eeq
373:
374: between the coupling
375: constants of the two theories. The physical interpretation of this
376: relation is discussed in \cite{Ambjorn:1999fp,Ambjorn:1998xu}: One obtains the CDT model
377: by chopping away all baby-universes from the Liouville gravity
378: theory, i.e.\ universes connected to the ``parent-universe'' by a
379: worm-hole of cut-off scale, and this produces the relation
380: \rf{6.1} \footnote{The relation \rf{6.1} is similar to the one
381: encountered in regularized bosonic string theory in dimensions
382: $d\geq 2$ \cite{Durhuus:1984in,Ambjorn:1985az,Ambjorn:1987wu}: The world sheet degenerates into
383: so-called branched polymer. The two-point function of these
384: branched polymers is related to the ordinary two-point function of
385: the free relativistic particle by chopping off (i.e.\ integrating
386: out) the branches, just leaving for each branched polymer
387: connecting two points in target space one {\it path} connecting
388: the two points. The mass-parameter of the particle is then related
389: to the corresponding parameter in the partition function for the
390: branched polymers as $X/\SL$ to $\tX/\sqrt{\tL}$ in eq.\ \rf{6.1}.
391: }. It is seen that $X \to \SL$ corresponds precisely to $\tilde{X}
392: \to \sqrt{\tL}/2$.
393:
394: While the starting point of the CDT quantization was the desire to
395: include only Lorentzian, causal geometries in the path integral,
396: the result \rf{2.a11} shows that after rotation to Euclidean
397: signature this prescription is in a natural correspondence with
398: the Euclidean Hartle-Hawking no-boundary condition, since all of
399: the geometries \rf{2.a11} have a continuation to $t \equ -t_0$,
400: where the spacetime is regular. It would be interesting if this
401: could be promoted to a general principle also in higher
402: dimensions. The computer simulations reported in
403: \cite{Ambjorn:2004qm,Ambjorn:2004pw,Ambjorn:2005db,Ambjorn:2005qt,Ambjorn:2005jj,Ambjorn:2006jf}
404: seem in accordance with this possibility.
405:
406: %The idea of CDT, i.e.\ quantum gravity defined via causal
407: %dynamical triangulations, is two-fold: firstly, inspired by
408: %Teitelboim \cite{teitelboim}, we insist, starting in a space-time
409: %with a Lorentzian signature, that only causal histories contribute
410: %to the quantum gravity path integral, and secondly, we assume a
411: %global time-foliation.
412:
413: %``Dynamical triangulation'' (DT) provides a simple regularization
414: %of the sum over {\it geometries} by providing a grid of piecewise
415: %linear geometries constructed from building blocks
416: %($d$-dimensional simplices if we want to construct a
417: %$d$-dimensional geometry, see \cite{book,leshouches} for reviews).
418: %The ultraviolet cut-off is the length of the side of the building
419: %blocks. CDT uses DT as the regularization of the path integral
420: %(see \cite{al,ajl5} for detailed descriptions of which causal
421: %geometries are included in the grid).
422:
423: %In two dimensions it is natural to study the proper-time
424: %``propagator'', i.e.\ the amplitude for two space-like boundaries
425: %to be separated a proper time (or geodesic distance) $T$. While
426: %this is a somewhat special amplitude, it has the virtue that other
427: %amplitudes, like the disc amplitude or the cylinder amplitude, can
428: %be calculated if we know the proper-time propagator
429: %\cite{kawai0,kn,gk,al}. When the path integral representation of
430: %this propagator is defined using CDT we can further, for each
431: %causal piecewise linear Lorentzian geometry, make an explicit
432: %rotation to a related Euclidean geometry. After this rotation we
433: %perform the sum over geometries in the this Euclidean regime.
434: %This sum is now different from the full Euclidean sum over
435: %geometries, leading to an alternative quantization of 2d quantum
436: %gravity (CDT). Eventually we can perform a rotation back from
437: %Euclidean proper time to Lorentzian proper time in the propagator
438: %if needed.
439:
440:
441: %In the following we will use continuum notation. A derivation of
442: %the continuum expressions from the regularized (lattice)
443: %expressions can be found in \cite{al}. We assume space-time has
444: %the topology $S^1\times [0,1]$, The action (rotated to Euclidean
445: %space-time) is: \beq\label{2.a} S[g] = \La \int \int \d x \d t
446: %\sqrt{g(x,t) } + X \oint \d l_1 + Y \oint \d l_2, \eeq where $\La$
447: %is the cosmological constant, $X,Y$ are two boundary cosmological
448: %constants, $g$ is a metric describing a geometry of the kind
449: %mentioned above, and the line integrals refer to the length of the
450: %boundaries, induced by $g$. The propagator $G_\La(X,Y;T) $ is
451: %defined by \beq\label{2.a0} G_\La (X,Y;T) = \int \cD [g] \;
452: %e^{-S[g]}, \eeq where the functional integration is over all
453: %``causal'' geometries $[g]$ such that the ``exit'' boundary with
454: %boundary cosmological constant $Y$ is separated a geodesic
455: %distance $T$ from the ``entry'' boundary with boundary
456: %cosmological constant $X$. As shown in \cite{al}, calculating the
457: %path integral \rf{2.a0} using the CDT regularization and taking
458: %the continuum limit where the side-length $a$ of the simplices
459: %goes to zero leads to the following expression\footnote{The
460: %asymmetry between $X$ and $Y$ is just due to the convention that
461: %the entrance boundary contains a marked point. Symmetric
462: %expressions where the the boundaries have no marked points or both
463: %have marked points can be found in \cite{alnr}}: \beq\label{2.a3}
464: %G_\La (X,Y;T) = \frac{\bX^2(T,X) -\La}{X^2-\La} \;
465: %\frac{1}{\bX(T,X) +Y}, \eeq where $\bX(T,X) $ is the solution of
466: %\beq\label{2.3} \frac{\d \bX}{\d T} = -(\bX^2-\La),~~~\bX(0,X) =X,
467: %\eeq or \beq\label{2.4} \bX(t,X) = \SL \coth \SL(t+t_0),~~~~X=\SL
468: %\coth \SL \,t_0. \eeq Viewing $G_\La(X,Y;T) $ as a propagator,
469: %$\bX(T) $ can be viewed as a ``runing'' boundary cosmological
470: %constant, $T$ being the scale. If $X > - \SL$ then $\bX(T) \to
471: %\SL$ for $T \to \infty$, $\SL$ being a ``fixed point'' (a zero of
472: %the ``$\b$-function'' $-(\bX^2-\La) $ in eq.\ \rf{2.3}).
473:
474: %Let $L_1$ denote the length of the entry boundary and $L_2$ the
475: %length of the exit boundary. Rather than consider a situation
476: %where the boundary cosmological constant $X$ is fixed we can
477: %consider $L_1$ as fixed. We denote the corresponding propagator
478: %$G_\La (L_1,Y;T) $. Similarly we can define $G_\La(X,L_2;T) $ and
479: %$G_\La(L_1,L_2;T) $. They are related by Laplace transformations.
480: %For instance: \beq\label{2.a5} G_\La(X,Y;T) = \int_0^\infty \d L_2
481: %\int_0^\infty \d L_1\; G(L_1,L_2;T) \;\e^{-XL_1-YL_2}. \eeq and
482: %one has the following composition rule for the propagator:
483: %\beq\label{2.a6} G_\La (X,Y;T_1+T_2) = \int_0^\infty \d L \; G_\La
484: %(X,L;T_1) \,G(L,Y,T_2). \eeq
485: