1: \documentclass[showpacs,amsmath,amssymb,twocolumn,nofootinbib]{revtex4}
2: \usepackage{epsfig}
3: \input{epsf}
4: %\documentclass{article}
5: %\input{epsf}
6: \usepackage{amssymb}
7: %\usepackage{graphicx}
8: \mathchardef\SGamma="7100
9: \begin{document}
10: \title{Thermodynamics via Creation from Nothing: Limiting
11: the Cosmological Constant Landscape\footnote{\normalsize To the
12: memory of Bryce DeWitt.}}
13:
14:
15: \author{A.O.Barvinsky}
16: \affiliation{Theory Department, Lebedev
17: Physics Institute, Leninsky Prospect 53, 119991 Moscow, Russia}
18: \author{A.Yu.Kamenshchik}
19: \affiliation{Dipartimento di Fisica and INFN, Via
20: Irnerio 46, 40126 Bologna, Italy\\
21: L.D. Landau Institute for Theoretical Physics, Kosygin str. 2,
22: 119334 Moscow, Russia}
23: \begin{abstract}
24: The creation of a quantum Universe is described by a {\em density
25: matrix} which yields an ensemble of universes with the cosmological
26: constant limited to a bounded range $\Lambda_{\rm min}\leq \Lambda
27: \leq \Lambda_{\rm max}$. The domain $\Lambda<\Lambda_{\rm min}$ is
28: ruled out by a cosmological bootstrap requirement (the
29: self-consistent back reaction of hot matter). The upper cutoff
30: results from the quantum effects of vacuum energy and the conformal
31: anomaly mediated by a special ghost-avoidance renormalization. The
32: cutoff $\Lambda_{\rm max}$ establishes a new quantum scale -- the
33: accumulation point of an infinite sequence of garland-type
34: instantons. The dependence of the cosmological constant range on
35: particle phenomenology suggests a possible dynamical selection
36: mechanism for the landscape of string vacua.
37: \end{abstract}
38: %\pacs{04.60.-m, 98.80.Qc, 98.80.Bp}
39: \maketitle
40:
41: Quantum cosmology \cite{Bryce,VilNB} and Euclidean quantum gravity
42: \cite{HH} might effectively restrict the landscape of string vacua.
43: This landscape is too big \cite{BoussoPolchinsky} to predict either
44: the observed particle phenomenology or large-scale structure
45: formation within string theory itself. Other methods have to be
46: invoked, at least some of them based on the cosmological
47: wavefunction \cite{OoguriVafaVerlinde,Tye,Brustein}. This is
48: dominated by the exponentiated Euclidean action, $\exp(-S_{\rm E})$,
49: calculated on the gravitational instanton which is a saddle point of
50: an underlying path integral over Euclidean 4-geometries. This
51: instanton gives rise to Lorentzian signature spacetime by analytic
52: continuation across minimal hypersurfaces. The continuation can be
53: interpreted either as quantum tunneling or as the creation of the
54: Universe from ``nothing". Thus, the most probable vacua of the
55: landscape become weighted by the minima of $S_{\rm E}$. This might
56: serve as a method of selecting a vacuum from the enormously big
57: string landscape.
58:
59: An immediate difficulty with this program arises from the infrared
60: catastrophe of small cosmological constant $\Lambda$. The
61: Hartle-Hawking wave function \cite{HH}, which describes nucleation
62: of the de Sitter Universe from the Euclidean 4-dimensional
63: hemisphere, has the form
64: \begin{eqnarray}
65: \Psi_{\rm HH}\sim \exp(-S_{\rm E})
66: =\exp(3\pi /2G\Lambda). \label{1}
67: \end{eqnarray}
68: This diverges for $\Lambda\to 0$ because of unboundedness of the
69: Euclidean gravitational action. Despite some early attempts to
70: interpret it as the origin of a zero value of $\Lambda$
71: \cite{bigfix}, this result remains both controversial and
72: anti-intuitive because it disfavors inflation and prefers creation
73: of infinitely large universes. Apart from the tunneling proposals of
74: \cite{tunnel} which employ an opposite sign in the exponential of
75: (\ref{1}) and thus open the possibility for opposite conclusions
76: \cite{scale}, no convincing resolution of this problem has thus far
77: been suggested.
78:
79: In this Letter we show that Euclidean path integration framework
80: naturally avoids this infrared catastrophe. We attain this result
81: by: i) extending the notion of Hartle-Hawking {\em pure} state to a
82: density matrix which describes a {\em mixed} quantum state of the
83: Universe and ii) incorporating the nonperturbative back reaction of
84: hot quantum matter on the instanton background \cite{slih}. These
85: extensions seem natural because whether the initial state of the
86: Universe is pure or mixed is a dynamical question rather than a
87: postulate. We address this question below by embedding both types of
88: states into a unified framework of a density matrix.
89: \begin{figure}[h]
90: \centerline{\epsfxsize 4.4cm \epsfbox{hh5.eps}} \caption{\small
91: Density matrix instanton. Dashed lines depict the Lorentzian
92: Universe nucleating at minimal surfaces $\Sigma$ and $\Sigma'$.
93: \label{Fig.1}}
94: \end{figure}
95:
96: A density matrix $\rho[\,\varphi,\varphi']$ is represented in
97: Euclidean quantum gravity \cite{Page} by an instanton having two
98: disjoint boundaries $\Sigma$ and $\Sigma'$ associated with its two
99: entries $\varphi$ and $\varphi'$ (collecting both gravity and matter
100: observables). The instanton interpolates between these, thus
101: establishing mixing correlations, see Fig.\ref{Fig.1}. In contrast,
102: for the density matrix of the pure Hartle-Hawking state the bridge
103: between $\Sigma$ and $\Sigma'$ is broken, so that the instanton is a
104: union of two disjoint hemispheres which smoothly close up at their
105: poles (Fig.\ref{Fig.2}) --- a picture illustrating the factorization
106: of $\hat\rho=|\Psi_{\rm HH}\rangle\langle\Psi_{\rm HH}|$.
107: \begin{figure}[h]
108: \centerline{\epsfxsize 4.3cm \epsfbox{hh1.eps}} \caption{\small
109: Density matrix of the pure Hartle-Hawking state represented by the
110: union of two vacuum instantons. \label{Fig.2}}
111: \end{figure}
112:
113: The main effect that we advocate here is that thermal fluctuations
114: and quantum conformal anomaly destroy the Hartle-Hawking instanton
115: and replace it with one filled by radiation. This is already
116: manifest in classical theory, specifically in the Euclidean
117: Friedmann equation for a scale factor $a(\tau)$,
118: $\dot{a}^2/a^2=1/a^2 - H^2 -C/a^4$ (we use spatially closed FRW
119: metric $ds^2 = N^2(\tau)\,d\tau^2 +a^2(\tau)\,d^2\Omega^{(3)}$ in
120: the gauge $N=1$ and express $\Lambda=3H^2$ in terms of the Hubble
121: constant). The radiation density $C/a^4$ prevents the
122: half-instantons from closing and allows $a$ to vary between two
123: turning points \cite{Halliwell,Rubakov}
124: $a_\pm=(1/\sqrt{2}H)(1\pm\sqrt{1-4CH^2})^{1/2}$.
125: This forces a tubular structure on the instanton which spans (at
126: least) one period of oscillation between $a_\pm$, provided the
127: constant $C$ characterizing the amount of radiation satisfies the
128: bound $4H^2C \leq 1$.
129:
130: The existence of radiation naturally follows from the partition
131: function of this state associated with the toroidal instanton
132: obtained by identifying $\Sigma'$ and $\Sigma$. The radiation back
133: reaction supports the instanton geometry in which it exists.
134: Remarkably, when the vacuum energy and conformal anomaly are taken
135: into account this bootstrap yields a set of instantons -- a
136: landscape -- only in the bounded range of $\Lambda$,
137: \begin{eqnarray}
138: \Lambda_{\rm min}<\Lambda<\Lambda_{\rm max}. \label{2}
139: \end{eqnarray}
140: All values $\Lambda<\Lambda_{\rm min}$ are completely eliminated
141: either because of the absence of instanton solutions or because of
142: their {\em infinitely large positive} action. A similar situation
143: holds for $\Lambda>\Lambda_{\rm max}$ -- no instantons exist there,
144: and the Lorentzian configurations in this overbarrier domain (if
145: any) are exponentially suppressed relative to those of (\ref{2}).
146:
147: To quantify the above picture consider the density matrix given by
148: the Euclidean path integral \cite{Page}
149: \begin{eqnarray}
150: \rho[\,\varphi,\varphi'\,]=
151: \mbox{\large$e$}^{\,\textstyle \varGamma}
152: %\!\!\!\!\!\!\!\!\!\!\!\!\!\!
153: \int
154: %\limits_{\,\,\,\,\,\,\,g,\,
155: %\phi\,|_{\,\Sigma,\Sigma'}\,=\,(\,\varphi,\,\varphi')}
156: %\!\!\!\!\!\!\!\!\!\!\!
157: D[\,g,\phi\,]\,
158: \exp\big(-S_{\rm E}[\,g,\phi\,]\big), \label{rho}
159: \end{eqnarray}
160: where $S_{\rm E}[\,g,\phi\,]$ is the classical action, and the
161: integration runs over gravitational $g$ and matter $\phi$ fields
162: interpolating between $\varphi$ and $\varphi'$ at $\Sigma$ and
163: $\Sigma'$. The statistical sum $\exp(-\varGamma)$ is given by a
164: similar path integral over periodic fields on the torus with
165: identified boundaries $\Sigma$ and $\Sigma'$.
166:
167: The back reaction follows from decomposing $[g,\phi]$ into a
168: minisuperspace $g_0(\tau)=\big(a(\tau),N(\tau)\big)$, and the
169: "matter" sector which includes also inhomogeneous metric
170: perturbations on minisuperspace background
171: $\varPhi(x)=(\phi(x),\psi(x),A_\mu(x), h_{\mu\nu}(x),...)$. With a
172: relevant decomposition of the measure $D[\,g,\phi\,]=Dg_0(\tau)
173: \times D\varPhi(x)$, the integral for $\varGamma$ expresses in terms
174: of the effective action $\varGamma[g_0(\tau)]$ of quantized matter
175: on the background $g_0(\tau)$,
176: $\exp(-\varGamma[g_0])=
177: \int D\varPhi(x)\,
178: \exp(\!-S_{\rm E}[g_0,\varPhi(x)])$,
179: as
180: \begin{eqnarray}
181: \mbox{\large$e$}^{\textstyle
182: -\SGamma}=
183: \int Dg_0(\tau)\,
184: \exp\Big(\!-\varGamma[g_0(\tau)]\Big). \label{intg0}
185: \end{eqnarray}
186: Our approximation will be to consider $\varGamma[g_0(\tau)]$ in the
187: one-loop order, $\varGamma[g_0]=S_{\rm E}[g_0]+\varGamma_{\rm
188: 1-loop}[g_0]$, and handle (\ref{intg0}) at the tree level, which is
189: equivalent to solving the {\em effective equations} for
190: $\varGamma[g_0]$.
191:
192: Remarkably, $\varGamma[\,g_0\,]$ is {\em exactly} calculable for
193: confor\-mally-invariant fields by a conformal transformation
194: \cite{FHH} relating generic FRW metric $ds^2 = a^2 d\bar s^2$ to
195: that of a static universe of a unit size, $d\bar s^2 = d\eta^2 +
196: d^2\Omega^{(3)}$ (these metrics are denoted below as $g$ and $\bar
197: g$, while $\eta$ is the conformal time). The total action reads
198: \begin{eqnarray}
199: &&\varGamma[\,a(\tau),N(\tau)\,]=
200: 2 \int_{\tau_-}^{\tau_+} d\tau\Big(\!-\frac{a\dot{a}^2}N
201: -Na+N H^2 a^3\Big)\nonumber\\
202: &&\qquad\qquad
203: +2B \int_{\tau_-}^{\tau_+}
204: d\tau \Big(\frac{\dot{a}^2}{Na}
205: -\frac16\,\frac{\dot{a}^4}{N^3 a}\Big)\nonumber\\
206: &&\qquad\qquad
207: + B \int_{\tau_-}^{\tau_+}
208: d\tau\,N/a+F\Big(2\int_{\tau_-}^{\tau_+}
209: d\tau\,N/a\Big)\,. \label{Gamma}
210: \end{eqnarray}
211: We work in units of $m_P=\sqrt{3\pi/4G}$, and the integration runs
212: between two turning points at $\tau_\pm$. The first line is the
213: classical part, the second line is the conformal contribution and
214: the last line is the one-loop action on the static instanton of the
215: metric $\bar g$.
216:
217: The conformal contribution $\varGamma_{\rm 1-loop}[g]-\varGamma_{\rm
218: 1-loop}[\bar g]$ is determined by the coefficients of $\Box R$, the
219: Gauss-Bonnet invariant $E = R_{\mu\nu\alpha\gamma}^2 -4R_{\mu\nu}^2
220: + R^2$ and Weyl tensor term in the conformal anomaly
221: $g_{\mu\nu}\delta
222: \SGamma_{\rm 1-loop}/\delta g_{\mu\nu} =
223: g^{1/2}
224: (\alpha \Box R +
225: \beta E + \gamma C_{\mu\nu\alpha\beta}^2)/4(4\pi)^2 $.
226: Specifically this contribution can be obtained by the technique of
227: \cite{TseytlinconfBMZ}; it contains higher-derivative terms $\sim
228: \ddot a^2$ which produce ghost instabilities in solutions of
229: effective equations. However, such terms are proportional to the
230: coefficient $\alpha$ which can be put to zero by adding the
231: following finite {\em local} counterterm
232: \begin{eqnarray}
233: \varGamma_{R}[g]
234: =\varGamma_{\rm 1-loop}[g]
235: +\frac1{2(4\pi)^2}
236: \frac\alpha{12}\int d^4x\,
237: g^{1/2}R^2(g). \label{renormalization}
238: \end{eqnarray}
239: This ghost-avoidance renormalization is justified by the requirement
240: of consistency of the theory at the quantum level. The contribution
241: $\varGamma_{R}[g]-\varGamma_{R}[\bar g]$ to the {\em renormalized}
242: action then gives the second line of (\ref{Gamma}) with
243: $B=3\beta/4$.
244:
245: The static instanton with a period $\eta_0$ playing the role of
246: inverse temperature contributes $\varGamma_{\rm 1-loop}[\bar g]
247: =E_0\,\eta_0+F(\eta_0)$, where the vacuum energy $E_0$ and free
248: energy $F(\eta_0)$ are the typical boson and fermion sums over field
249: oscillators with energies $\omega$ on a unit 3-sphere
250: %\begin{equation}
251: $E_0=\pm\sum_{\omega}
252: \omega/2\,,\,\,\,F(\eta_0)=\pm\sum_{\omega}
253: \ln\big(1\mp e^{-\omega\eta_0}\big)$.
254: %\end{equation}
255: The renormalization (\ref{renormalization}) which should be applied
256: also to $\varGamma_{\rm 1-loop}[\bar g]$ modifies $E_0$, so that
257: $\varGamma_{R}[\bar g] =C_0\,\eta_0+F(\eta_0)$, $C_0\equiv
258: E_0+3\alpha/16$. This gives the third line of Eq.(\ref{Gamma}) with
259: $C_0=B/2$. This universal relation between $C_0$ and $B=3\beta/4$
260: follows from the known anomaly coefficients \cite{confanomaly} and
261: the UV-renormalized Casimir energy in a static universe \cite{E_0}
262: for scalar, Weyl spinor and vector fields respectively having:
263: \begin{eqnarray}
264: \alpha=\frac1{90}\!\times\!\left\{\begin{array}{c} \!-1 \\
265: \!-3\\
266: \!18\end{array}\right.\!\!,
267: \,\,
268: \beta=\frac1{360}\!\times\!
269: \left\{\begin{array}{cl} \!\!2\\
270: \!\!11\\
271: \!\!124\end{array}\right.\!\!\!\!,
272: \,\,
273: E_0=\frac1{960}\!\times\!\left\{\begin{array}{c} \!\!4 \\
274: \!17\\
275: \!88\end{array}\right.\!\!\!. \label{5000}
276: \end{eqnarray}
277: It is important that for conformally invariant fields the nonlocal
278: action (\ref{Gamma}) is exact, and contains no other terms of higher
279: order in the curvature.
280:
281: The effective equation $\delta\varGamma/\delta N(\tau)=0$ has the
282: form of the classical equation modified by the quantum $B$-term
283: \begin{eqnarray}
284: &&\frac{\dot{a}^2}{a^2}
285: +B \left(\frac12\,\frac{\dot{a}^4}{a^4}
286: -\frac{\dot{a}^2}{a^4}\right) =
287: \frac{1}{a^2} - H^2 -\frac{C}{ a^4}, \label{efeq}\\
288: %\end{eqnarray}
289: %\begin{eqnarray}
290: &&C = B/2 +F'(\eta_0),\,\,
291: \eta_0 = 2\int_{\tau_-}^{\tau_+}
292: d\tau/a(\tau). \label{bootstrap}
293: \end{eqnarray}
294: Remarkably, the contribution of the nonlocal $F(\eta_0)$ in
295: (\ref{Gamma}) reduces to the radiation {\em constant} $C$ as a {\em
296: nonlocal functional} of $a(\tau)$, determined by the {\em bootstrap}
297: equation (\ref{bootstrap}). Here $F'(\eta_0)\equiv
298: dF(\eta_0)/d\eta_0>0$ is the energy of a hot gas of particles, which
299: adds to their vacuum energy $B/2$.
300:
301: \begin{figure}[h]
302: \centerline{\epsfxsize 8.5cm \epsfbox{hh10.eps}} \caption{\small
303: Instanton domain in the $(H^2,C)$-plane. Garland families are shown
304: for $k=1,2,3,4$. Their sequence accumulates at the critical point
305: $(1/2B,B/2)$.
306: \label{Fig.4}}
307: \end{figure}
308: Periodic instanton solutions of Eqs.(\ref{efeq})-(\ref{bootstrap})
309: exist only inside the curvilinear wedge of $(H^2,C)$-plane between
310: bold segments of the upper hyperbolic boundary and the lower
311: straight line boundary of Fig.\ref{Fig.4},
312: \begin{equation}
313: 4CH^2\leq 1,\,\,\,C \geq B-B^2 H^2,
314: \,\,\,B H^2\leq 1/2. \label{restriction1}
315: \end{equation}
316: Below this domain the solutions are either complex and aperiodic or
317: suppressed by {\em infinite positive} Euclidean action. Above this
318: domain only Lorentzian (overbarrier) configurations exist, but they
319: are again exponentially damped relative to instantons in
320: (\ref{restriction1}).
321:
322: These properties are based on the fact that the turning points of
323: (\ref{efeq}) exactly coincide with classical $a_\pm$, but $a_-$
324: exists only when $a_-^2 \geq B$, which gives rise to
325: (\ref{restriction1}). Otherwise, $a(\tau)$ at the contraction phase
326: becomes complex or runs to zero which violates instanton
327: periodicity. In the latter case a smooth Hartle-Hawking instanton
328: with $a_-=0$ forms and yields $\eta_0\to\infty$ in view of
329: (\ref{bootstrap}), so that $F(\eta_0)\sim F'(\eta_0)\to 0$.
330: Therefore, its on-shell action
331: \begin{equation}
332: \varGamma_0= F(\eta_0)\!-\!\eta_0 F'(\eta_0)
333: +4\!\int_{a_-}^{a_+}\!
334: \frac{da \dot{a}}{a}\Big(B-a^2
335: -\frac{B\dot{a}^2}{3}\Big) \label{action-instanton}
336: \end{equation}
337: due to $B>0$ diverges to $+\infty$ at $a_-=0$ and completely rules
338: out pure-state instantons \cite{slih}.
339:
340: Moreover, inside the range (\ref{restriction1}) our bootstrap
341: eliminates the infrared catastrophe of $\Lambda\to 0$. Indeed
342: $\eta_0\to\infty$ as $H^2\to 0$, so that due to (\ref{bootstrap})
343: $C\to B/2$, but this is impossible because in view of
344: (\ref{restriction1}) $C\geq B$ at $H^2=0$. Thus, instanton family
345: never hits the $C$-axes of $H^2=0$ and can only interpolate between
346: the points on the boundaries of the domain (\ref{restriction1}). For
347: a conformal scalar field the numerical analysis gives such a family
348: \cite{slih} starting at $H^2\approx2.00,\,\,C\approx0.004,\,\,\,
349: \varGamma_0\approx-0.16$, and terminating at
350: $H^2\approx13.0,\,\,C\approx0.02 ,\,\,\,\varGamma_0\approx-0.09$.
351: The upper point describes the static universe filled by a hot
352: radiation with the temperature $T=H/\pi\sqrt{1-2BH^2}$, whereas the
353: lower point establishes the lower bound of the $\Lambda$-range.
354:
355: The upper bound of the landscape follows from the existence of {\em
356: garlands} that can be obtained by glueing together into a torus $k$
357: copies of a simple instanton \cite{Halliwell,mult-inst}; see
358: Fig.\ref{Fig.5}. Their formalism is the same as above except that
359: the conformal time in (\ref{bootstrap}) and the integral term of
360: (\ref{action-instanton}) should be multiplied by $k$.
361: \begin{figure}[h]
362: \centerline{\epsfxsize 5cm \epsfbox{hh11.eps}} \caption{\small The
363: garland segment consisting of three folds of a simple instanton
364: glued at surfaces of a maximal scale factor.
365: \label{Fig.5}}
366: \end{figure}
367: As in the case of $k=1$, garland families interpolate between the
368: lower and upper boundaries of (\ref{restriction1}). They exist for
369: all $k$, $1\leq k\leq\infty$, and their infinite sequence
370: accumulates at the critical point $C=B/2$, $H^2=1/2B$, where these
371: boundaries merge. Within the $1/k^2$-accuracy the upper and lower
372: points of each family coincide and read
373: \begin{eqnarray}
374: H^2_{(k)}\simeq \frac1{2B}
375: \left(1 - \frac{\ln^2k^2}{2k^2\pi^2}
376: \right),\,\,\,\varGamma_0^{(k)}\simeq
377: -B\,\frac{\ln^3 k^2}{4k^2\pi^2}. \label{action-k}
378: \end{eqnarray}
379: With a growing $k$, garlands become more and more static and cold
380: with $T_{(k)}\simeq 1/(\sqrt B \ln k^2)\to 0$. Contrary to
381: \cite{mult-inst} the garland action is not additive in $k$, so that
382: as $k\to\infty$ garlands do not dominate the ensemble. Their
383: sequence converges to the instanton with $H^2_{\rm max}=1/2B$, which
384: gives the upper bound of the range (\ref{2}).
385:
386: Thus, our Universe is created in a hot mixed state, but its
387: evolution does not contradict the large-scale structure formation.
388: After nucleation from the instanton the Universe expands; its
389: radiation dilutes, so that $\Lambda$ starts dominating and generates
390: inflation under an assumption that everywhere above $\Lambda$ is a
391: composite field (like an inflaton) decaying at the exit by a
392: standard slow-roll scenario.
393:
394: The ensemble of universes belongs to a bounded range (\ref{2}) of
395: $\Lambda=3H^2$. Its infrared cutoff is provided by the radiation
396: back reaction and survives even in the classical limit as $B\to 0$.
397: In contrast, the high-energy cutoff at
398: \begin{eqnarray}
399: \Lambda_{\rm max}=3m_P^2/2B,\,\,\,
400: m_P^2\equiv 3\pi/4G, \label{top}
401: \end{eqnarray}
402: is the quantum effect of vacuum energy and the conformal anomaly;
403: this generates a new scale in gravity theory.
404:
405: We have considered only conformal fields which make our model
406: exactly solvable and provide critically important positivity of the
407: constant $B=3\beta/4$, cf. (\ref{5000}). Moreover, conformal
408: invariance together with the ghost-avoidance renormalization renders
409: a particular value of the vacuum energy $B/2$ in (\ref{bootstrap})
410: which yields the upper boundary of (\ref{2}) exactly at the critical
411: point $(1/2B,B/2)$ of Fig.\ref{Fig.4}. Even if non-conformal fields
412: qualitatively preserve the whole picture, they are likely to break
413: this relation. Then if $C_0<B/2$ all garlands survive, though they
414: saturate at $\Lambda_{\rm max}$ with a finite temperature. If
415: $B>C_0>B/2$, their sequence is truncated at some $k$. Finally, if
416: $C_0>B$ the infrared catastrophe occurs again -- the $k=1$ instanton
417: family hits the $C$-axes at $C_0$.
418:
419: Conformal invariance can be justified as a good approximation when
420: conformal particles outnumber non-conformal ones. Moreover, their
421: large number $N$ justifies a semiclassical expansion by scaling down
422: the range (\ref{2}). Indeed, for a single scalar field the latter is
423: determined by Planckian values, $\Lambda_{\rm min}\approx 8.99\,
424: m_P^2$, $\Lambda_{\rm max}=360\, m_P^2$ which, however, decrease as
425: $1/N$ in view of the simple scaling $C\to NC$, $B\to NB$,
426: $F(\eta_0)\to NF(\eta_0)$ and $H^2\to H^2/N$. Semiclassical
427: expansion can also be justified for large $B=3\beta/4$ growing with
428: spin, cf.(\ref{5000}), because the domain (\ref{2}) with (\ref{top})
429: shrinks to a narrow subplanckian range when ascending the particle
430: hierarchy.
431:
432: Though motivated by the string landscape, all the above results hold
433: outside of the string theory context and, as a feedback, suggest a
434: long-sought selection mechanism for the plethora of string vacua.
435: Modulo the details of a relevant $4D$-compactification, this might
436: work as follows. For $B$ growing with $N$ and spin, the upper scale
437: (\ref{top}) decreases towards the increasing phenomenology scale,
438: and approaches the latter at the string scale $m_{\rm s}^2$ where a
439: positive $\Lambda$ might be generated by the mechanism of
440: \cite{KKLT}. Our conjecture is that at this scale our bootstrap
441: becomes perturbatively consistent, provided $m_P^2/B\simeq m_{\rm
442: s}^2\ll m_P^2$, and selects from the string landscape a small subset
443: compatible with observations.
444:
445: Our results hold within the Euclidean path integral (\ref{rho})
446: which automatically excludes Lorentzian configurations possibly
447: existing above the upper boundary of (\ref{restriction1}),
448: $4CH^2>1$. However, one can imagine an extended formulation of
449: quantum gravity generalizing (\ref{rho}) to a wider path integration
450: domain. Our conclusions nevertheless remain true. Indeed the
451: effective action scales as $\varGamma_0\!\sim\!-\sqrt B$, $B\gg 1$,
452: and because it is {\em negative} our landscape at the scale $m_{\rm
453: s}$ is weighted by $\exp(\#\sqrt B)\!\simeq\!\exp(\#\,m_P/m_{\rm
454: s})\gg 1$. Therefore it strongly dominates over Lorentzian
455: configurations, the amplitudes of the latter being $O(1)$ in view of
456: their pure phase nature. Thus, our results look robust against
457: possible generalizations of Euclidean quantum gravity.\\
458:
459:
460: A.B. thanks H.Tye, C.Deffayet, J.Khoury, R.Woodard, T.Osborn and
461: especially Jim Hartle for thought provoking discussions and helpful
462: suggestions. A.B. was supported by the RFBR grant 05-02-17661 and
463: the grant LSS 4401.2006.2. A.K. was partially supported by the RFBR
464: grant 05-02-17450 and the grant LSS 1757.2006.2.
465:
466: \begin{thebibliography}{99}
467: \bibitem{Bryce}B.S.DeWitt, Phys. Rev. {\bf 160}, 1113 (1967).
468: \bibitem{VilNB}A. Vilenkin, Phys. Lett. {\bf B 117}, 25 (1982);
469: Phys. Rev. {\bf D 27}, 2848 (1983).
470: \bibitem{HH}J.B.Hartle and S.W.Hawking, Phys.Rev. {\bf D28}, 2960
471: (1983); S.W.Hawking, Nucl. Phys. {\bf B 239}, 257 (1984).
472: \bibitem{BoussoPolchinsky}R.Bousso and J.Polchinsky, JHEP {\bf 06}
473: 006 (2000); M.R.Douglas, JHEP {\bf 05}, 046 (2003), hep-th/0303194.
474: \bibitem{OoguriVafaVerlinde}H.Ooguri et al,
475: %C.Vafa and E.Verlinde,
476: Lett. Math. Phys. {\bf 74}, 311 (2005).
477: \bibitem{Tye}H.Firouzjahi et al,
478: %S.Sarangi and S.-H.H.Tye,
479: JHEP {\bf 0409}, 060 (2004); S.Sarangi and S.-H.H.Tye,
480: hep-th/0505104, hep-th/0603237.
481: \bibitem{Brustein}R.Brustein and S.P.de Alwis, Phys. Rev. {\bf D
482: 73}, 046009 (2006).
483: \bibitem{bigfix}S.W.Hawking, Phys. Lett. {\bf B 134}, 403 (1984);
484: S.R.Cole\-man, Nucl. Phys. {\bf B 310}, 643 (1988).
485: \bibitem{tunnel}A.D. Linde, JETP {\bf 60}, 211 (1984);
486: A.Vilenkin, Phys. Rev. {\bf D 30}, 509 (1984).
487: \bibitem{scale}A.O. Barvinsky and A.Yu. Kamenshchik,
488: Class. Quantum Grav. {\bf 7}, L181 (1990); Phys. Lett. {\bf B 332},
489: 270 (1994).
490: %\bibitem{debate}A.Vilenkin, Phys. Rev. {\bf D58}, 067301 (1988),
491: %gr-qc/9804051; gr-qc/9812027.
492: \bibitem{slih}A.O.Barvinsky and A.Yu.Kamenshchik, %Cosmological
493: %landscape from nothing: some like it hot,
494: JCAP {\bf 09}, 014 (2006).
495: \bibitem{Page}D.N.Page, Phys. Rev. {\bf D 34}, 2267 (1986).
496: \bibitem{Halliwell}J.J. Halliwell and R.C.Myers, Phys. Rev. D 40, 4011
497: (1989).
498: \bibitem{Rubakov}G.V.Lavrelashvili et al,
499: %V.A.Rubakov, M.S.Serebryakov and P.G. Tinyakov,
500: Nucl. Phys. {\bf B 329}, 98 (1990); A.Yu.Kamenshchik et al,
501: %I.M.Khalatnikov and A.V.Toporensky,
502: Phys. Lett. {\bf B 357}, 36 (1995).
503: \bibitem{FHH}M.V.Fischetti, J.B.Hartle and B.L.Hu, Phys. Rev. {\bf D 20},
504: 1757 (1979).
505: \bibitem{TseytlinconfBMZ}E.S.Fradkin and A.A.Tseytlin,
506: Phys. Lett. {\bf B 134}, 187 (1984); A.O.Barvinsky et al,
507: %A.G.Mirzabekian and V.V.Zhytnikov,
508: gr-qc/9510037; P.O.Mazur and E.Mottola, Phys. Rev. {\bf D 64},
509: 104022 (2001).
510: \bibitem{confanomaly}J.S.Dowker and R.Critchley, Phys. Rev. {\bf
511: D 16}, 3390 (1977); S.M.Christensen, Phys. Rev. {\bf D 17}, 946
512: (1978).
513: \bibitem{E_0}J.S.Dowker and R.Critchley, J. Phys. {\bf A 9}, 535
514: (1976); S.G.Mamaev et al,
515: %V.M.Mo\-stepanenko and A.A.Starobinsky,
516: Sov. Phys. JETP, {\bf 43}, 823 (1976); L.H.Ford, Phys. Rev. {\bf D
517: 14}, 3304 (1976); P.Candelas and J.S.Dowker, Phys. Rev. {\bf D 19},
518: 2902 (1979).
519: \bibitem{mult-inst}W.Fischler,
520: D.Morgan and J.Polchinsky, Phys. Rev.{\bf D42} (1990) 4042.
521: \bibitem{KKLT}S.Kachru et al,
522: %R.Kallosh, A.Linde and S.P.Trivedi,
523: Phys.Rev. {\bf D 68}, 046005 (2003); S.Kachru et al,
524: %R.Kallosh, A.Linde, J.Maldacena, L.McAl\-lister and S.P.Trivedi,
525: JCAP {\bf 10}, 013 (2003).
526: \end{thebibliography}
527: \end{document}
528: