y claim 2.6 without loss of generality {\mathfrak b}_\lambda = {\mathfrak d}_\lambda = \mu. In particular, \lambda is super-compact and \boxdot_{<\lambda, \theta_{*}, \overline{\theta}} holds in the generic extension. Let \langle f^*_\alpha:\alpha <\mu\rangle witness {\mathfrak b}_\lambda = {\mathfrak d}_\lambda = \mu and without loss of generality \alpha < \beta < \mu \Rightarrow f^*_\alpha <_{J^{{\rm bd}}_\lambda} f^*_\beta.
Recalling Definitions 1.8, 1.9, Claim 1.10, Theorem 1.11, in \mathbf V there are \beta(*), \mathbf q,\bar u,{\mathscr U}_*,\ldots such that:
(*)_1(A) \mathbf q \in \mathbf Q_{\lambda,\bar\theta,\beta(*)} so in particular we have (in {\mathbf q}):
(a) \langle {\mathbb P}_{0,\alpha},\let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3Q\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}_{0,\beta}:
\alpha \le \beta(*),\beta <
\beta(*)\rangle is a (< \lambda)-support iteration,
(b) \bar u = \langle u_\beta:\beta <
\beta(*)\rangle,\bar{{\mathscr P}} = \langle {\mathscr P}_\beta:\beta < \beta(*)\rangle,
(c) u_\beta \subseteq \beta,{\mathscr P}_\beta \subseteq [u_\beta]^{\le \lambda} is closed under subsets,
(d) \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3Q\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}_{0,\beta} has generic \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}_\beta \in \prod\limits_{\varepsilon < \lambda} \theta_\varepsilon, and is (< \lambda)-strategically complete,
(e) \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3Q\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}_{0,\beta} is as in 1.8(e) so is \subseteq {\mathbb Q}_{\bar\theta}^{\mathbf V[\langle \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}_\alpha:\alpha \in u_\beta\rangle]} and \Vdash_{{\mathbb P}_{\beta +1}} ``\let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}_\beta \in \prod\limits_{\varepsilon < \lambda} \theta_\varepsilon" and \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth} = \langle \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}_\beta:\beta < \beta(*)\rangle,
(f) {\mathscr U}_* \subseteq \beta(*) has order type { \kappa } and \langle \beta^*_i:i \le \kappa\rangle lists {\mathscr U}_* \cup \{\beta(*)\} in increasing order, and \beta(*) = \sup({\mathscr U}_{*}),
(g) if \beta \in {\mathscr U}_* then {\mathscr U}_* \cap \beta \subseteq u_ \beta and [{\mathscr U}_ * \cap \beta ]^{\le \lambda} \subseteq {\mathscr P}_\beta and \Vdash_{{\mathbb P}_{0,\beta +1}} “if \nu \in \mathbf V[\langle \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}_\alpha: \alpha \in {\mathscr U}_* \cap \beta\rangle] \cap \prod\limits_{\varepsilon < \lambda} \theta_\varepsilon then \nu <_{J^{{\rm bd}}_\lambda} \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}_\beta",
(h) if \alpha \le \beta(*) then {\mathbb P}_{0,\alpha} is (< \lambda)-strategically complete, {(< \lambda,\theta _ *, \bar{\theta } ) }-directed complete and \lambda^+-c.c.,
(i) {{\mathbb P}}_{1, \alpha}, {{\mathbb P}}_{1, {\mathscr U}} are as in 1.9,
(j) \theta _* \ge (2^\mu )^+.
(B) letting {\mathbb P}'_i = {\mathbb P}_{\mathbf q,1,\{\beta^*_j:j <i\}} for i \le \kappa we have:
(a) The sequence \bar{ \mathbb{P} }'= \langle {\mathbb P}'_i:i \le { \kappa }\rangle of forcing notions is \lessdot-increasing, and is continuous for ordinals i \le { \kappa } of cofinality > \lambda see [Sh:1126, 2.5(8)=Lb14(8)], but the continuity will not be used,
(b) {\mathbb P}'_i is (< \lambda)-strategically complete for i \le { \kappa },
(c) (\prod\limits_{\varepsilon < \lambda} \theta_\varepsilon)^{\mathbf V[{\mathbb P}'_{{ \kappa }}]} = \cup\{(\prod\limits_{\varepsilon < \lambda} \theta_\varepsilon)^{\mathbf V[{\mathbb P}'_i]}:i < { \kappa }\},
(d) The sequence \langle {\mathbb P}_{1,\beta}:\beta \le \beta(*)\rangle is a sequence of forcing notions, \lessdot-increasing and if \beta \le \beta(*) then {\mathbb P}_{0,\beta} \lessdot {\mathbb P}_{1,\beta}, in fact is dense in it and if i \le { \kappa } then {\mathbb P}'_i \lessdot {\mathbb P}_{1,\beta^*_i}.
We shall mention more properties later.
[Why are there such objects? We apply 1.11 and 1.8 and 1.10, that is [Sh:1126]].
Also,
(*)_2
(a) recall \langle \beta^*_i:i \le { \kappa } \rangle lists {\mathscr U}_* \cup \{\beta(*)\} in increasing order,
(b) for i < { \kappa } let \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3g\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}'_i be \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}_{\beta^*_i} (to avoid excessive subscripts),
(c) let \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3g\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}' = \langle \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3g\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}'_i:i < \kappa\rangle,
(d) {\mathscr P}_\alpha = {\mathscr P}_{\mathbf q,\alpha} and without loss of generality u_\alpha = \cup\{u:u \in {\mathscr P}_\alpha\} for \alpha < \beta(*).
(*)_3 if u \in {\mathscr P}_\alpha,\alpha < \beta(*) then \Vdash_{{\mathbb P}_{0,\alpha +1}} ``\let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3 \crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth} _\alpha \in \prod\limits_{\varepsilon < \lambda} \theta_\varepsilon dominates (\prod\limits_{\varepsilon < \lambda} \theta_\varepsilon)^{\mathbf V[\langle \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3g\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}_\beta:\beta \in u \rangle]}", the order being modulo J^{\text{bd}}_\lambda.
[Why? By the choice of the forcing, see 1.4 or (*)_1(A)(g) above].
(*)_4 we have
(a) \Vdash_{{\mathbb P}'_\kappa} ``\let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3g\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}' =
\langle \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3g\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}'_i:i < \kappa\rangle is <_{J^{\text{bd}}_\lambda}-increasing and cofinal in (\prod_{\varepsilon < \lambda} \theta_\varepsilon, <_{J^{\text{bd}}_\lambda})".
(b) for every p \in {\mathbb P}_{\beta(*)} and \alpha \in {\rm dom}(p) \cap {\mathscr U}_{*}, for every large enough i < \kappa, we have p \Vdash_{{\mathbb P}_{0, \beta(\ast)}}“\let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3f\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}_{p(\alpha)} \leq \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3g\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}_{i}' = \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}_{\beta_{i}^{*}} \mod J_{\lambda}^{\rm{bd}}”,
(c) \Vdash_{{\mathbb P}_{0, \beta(*)}}“\overline{\let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3g\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}}' = \langle \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3g\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}_{i}': i < \kappa \rangle is <_{J_{\lambda}^{\rm{bd}}}-increasing and cofinal in \left( \Pi_{{\varepsilon} < \lambda \theta_{{\varepsilon}}} , <_{J_{\lambda}^{\rm{bd}}}\right)”.
[Why? Clause (a) holds by (*)_3 noting that (\prod\limits_{\varepsilon < \lambda} \theta_\varepsilon)^{\mathbf V[{\mathbb P}'_\kappa]} =
\cup\{(\prod\limits_{\varepsilon < \lambda}
\theta_\varepsilon)^{\mathbf V[{\mathbb P}'_i]}:i < \kappa\} which holds by 1.11(A)(d). Clause (b) holds by 1.11(A)(e) alternatively, in [Sh:1126] this notation means:
if \alpha \in {\mathscr U}_{*} and t \in L_{{\mathbf m}} \setminus M_{{\mathbf m}}, \alpha(i) \in u_{\alpha} \cap (t / E_{{\mathbf n}}') for i < \lambda and {\mathbf B} a \lambda-Borel function from {}^{\lambda} \left( \Pi_{{\varepsilon} < \lambda} \theta_{{\varepsilon}} \right) into \Pi_{{\varepsilon} < \lambda} \theta_{{\varepsilon}} then for every i < \kappa large enough \Vdash“{\mathbf B}(\dots, \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}_{\alpha({\varepsilon})}, \dots)_{i < \lambda} \leq_{J_{\lambda}^{\rm{bd}}} g'”.
.
Clause (c) holds by (*)_{1}(A)(g)].
Now,
(*)_5 \Vdash_{{\mathbb P}'_\kappa} “cov_\lambda(meagre) \le \kappa".
[Why? First, notice that we can look at \prod\limits_{\varepsilon < \lambda} \theta_\varepsilon instead of {}^\lambda 2.
Second, for each \varepsilon < \lambda,i < \kappa the set B_{\varepsilon,i} = \{\eta \in \prod\limits_{\xi < \lambda} \theta_\xi: for every \zeta \in [\varepsilon,\lambda) we have \eta(\zeta) \le \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3g\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}'_i(\zeta) < \theta_\zeta\} is closed nowhere dense, and by (*)_4 we have \mathbf V^{{\mathbb P}'_\kappa} \models ``\prod_{\zeta < \lambda} \theta_\zeta = \cup\{B_{\varepsilon,i}:\varepsilon < \lambda,i < \kappa\}". In fact, \langle B_{0,i}:i < \kappa\rangle suffice.
Alternatively we have \langle g'_i : i < \kappa \rangle is < _{J^{\rm bd}_\lambda }-increasing cofinal in \Pi _{\varepsilon < \lambda } \theta _ \varepsilon and let {{\mathscr W} } _{i, \zeta } := \{ \eta: \eta \in {}^ \lambda 2 and for every \varepsilon \in [ \zeta, \lambda ) we have either \eta \restriction [ \Sigma _{\xi < \varepsilon} \theta _ \xi , \Sigma _{\xi \le \varepsilon } \theta _ \xi) is constantly zero or {\rm min } \{ \alpha : \Sigma _{\xi < \varepsilon } \theta _ \xi + \alpha \in \eta ^{-1}(\{ 1 \} ) \} < g'_i (\varepsilon ) \}. So {{\mathscr W} } _{i, \zeta } is a closed nowhere dense subset of {}^ \lambda 2 and \cup \{ {{\mathscr W} } _{i, \zeta } : i < \kappa, \zeta < \lambda \} = {}^\lambda 2 and \kappa \times \lambda has cardinality \lambda + \kappa = \kappa because if f \in {}^{ \lambda } 2 then we define \nu _f \in \Pi _{\varepsilon < \lambda } \theta _ \varepsilon as follows: for \varepsilon < \lambda:
(a) if f {\restriction} [ \Sigma _{\xi < \varepsilon }\theta _ \xi, \Sigma _{\xi \le \varepsilon } \theta _ \varepsilon ) is not constantly zero then we let \nu _f (\varepsilon) = \min \{\alpha : f(\Sigma _{\xi < \varepsilon } \theta _ \varepsilon + \alpha ) = 1 \};
(b) if otherwise then let \nu_f (\varepsilon ) = 0.
So there are i < \kappa and \varepsilon < \lambda such that: \zeta \in [\varepsilon, \lambda ) \Rightarrow \nu _ f ( \zeta ) < g'_i(\zeta ). Now it is easy to check that f \in {\mathscr W } _{i, \varepsilon}.]
Lastly,
(*)_7 \Vdash_{{\mathbb P}'_\kappa}``{\rm cov}_\lambda({\rm meagre}) \ge
\kappa".
[Why? For i < \kappa let us define the {\mathbb P}'_{i +1}-name \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth} '_i of a member of {}^\lambda 2 by \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth} '_i(\varepsilon)=0 iff \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3g\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}'_i(\varepsilon) is even. Now clearly \Vdash_{{\mathbb P}'_{i +1}} ``\let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth} '_i is a \lambda-Cohen sequence over \mathbf V^{{\mathbb P}'_i}". (But let us elaborate; \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth} '_i is also a {\mathbb P}_{\beta^*_i+1}-name and \Vdash_{{\mathbb P}_{\beta^*_i+1}} ``\let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth} '_i is \lambda-Cohen over \mathbf V^{{\mathbb P}_{\beta^*_i}} hence over \mathbf V^{{\mathbb P}'_i}"; the last hence because {\mathbb P}'_i \lessdot {\mathbb P}_{1, \beta^*_i}. As {\mathbb P}_{\beta^*_i+1} \lessdot {\mathbb P}_{\beta^*_{i+1}} and {\mathbb P}'_{i+1} \lessdot {\mathbb P}_{\beta^*_{i+1}} we are done ).
Also every closed nowhere dense subset of {}^\lambda 2 from \mathbf V^{{\mathbb P}'_{{ \kappa }}} is from \mathbf V^{{\mathbb P}'_i} for some i < { \kappa }. So if p \Vdash ``{\rm cov}_\lambda({\rm meagre}) < \kappa" then for some \zeta < \kappa and \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3A\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}_{\varepsilon}({\varepsilon} < \zeta) we have p \Vdash ``\let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3A\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}_{\varepsilon} is a closed no-where dense subset of {}^\lambda 2 for \varepsilon < \zeta" and p \Vdash ``\bigcup\limits_{ \varepsilon < \zeta } \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3 A\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}_ \varepsilon is equal to the set of {}^\lambda 2". Without loss of generality each \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3A\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}_{\varepsilon} is a {\mathbb P}_{i({\varepsilon})}-name, i({\varepsilon}) < \kappa and recall that \kappa is regular. Hence i = \sup\{i({\varepsilon}):{\varepsilon} < \zeta\} < \kappa and \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3g\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth} '_i gives a contradiction to the choice of \langle \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3A\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}_{\varepsilon}:{\varepsilon} <
\zeta\rangle; so (*)_{6} holds indeed.]
The reader may look at some explanation in 2.9.
Now we come to the main and last point recalling \langle f^*_\alpha:\alpha < \mu\rangle from Claim 2.6
(*)_7 \Vdash_{{\mathbb P}'_{{ \kappa }}} “no \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3f\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth} \in ({}^\lambda \lambda) dominates \{f^*_\alpha:\alpha < \mu\}".
We shall show that it suffices to prove (*)_7 for proving Lemma 2.3(2), and then that (*)_7 holds, thus finishing.
Why it suffices? As \langle f^*_\alpha:\alpha < \mu\rangle is <_{J^{\text{bd}}_\lambda}-increasing and cf(\mu) = \mu > \lambda, this implies \Vdash_{{\mathbb P}'_\kappa} ``{\mathfrak d}_\lambda \ge \mu". Also in \mathbf V,\mu^\lambda = \mu > \kappa > \lambda and |{\mathbb P}'_{{ \kappa }}| = \kappa^\lambda by (A)(g) of 1.10(4) which is \le \mu and {\mathbb P}'_\kappa satisfies the \lambda^+-c.c. hence \Vdash_{{\mathbb P}'_\kappa} ``2^\lambda = \mu ", hence together \Vdash_{{\mathbb P}'_\kappa} ``{\mathfrak d}_\lambda = \mu". Also by (*)_1(B)(b), “{\mathbb P}'_{{ \kappa }} is (< \lambda)-strategically complete and \lambda^+-c.c." and by (*)_5 + (*)_6 we know that “{\rm cov}_\lambda({\rm meagre}) = \kappa" so we are done; hence (*)_7 is really the last piece missing. The rest of the proof is dedicated to proving that (*)_7 holds.
We shall use further nice properties of {\mathbb P}'_j,\let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3g\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}'_i (j \le { \kappa },i < { \kappa }) which hold by (*)_1 + (*)_2 (and (*)_3,(*)_4) and their proof, i.e. 1.10, 1.11 and see [Sh:1126, 2.12=Lb35, 2.13=Lb38].
\boxplus_1
(a)
(\alpha) \langle \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3g\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}'_\gamma:
\gamma < { \kappa }\rangle is generic for {\mathbb P}'_{{ \kappa }}, i.e., if \mathbf G is a subset of {\mathbb P}'_{{ \kappa }} generic over \mathbf V and g'_i = \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3g\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}'_i[\mathbf G] then \mathbf V[\mathbf G] = \mathbf V[\langle g'_i:i < { \kappa }\rangle],
(\beta) if in addition \nu \in ({}^\lambda \lambda)^{\mathbf V[\mathbf G]} then for some \rho \in ({}^\lambda { \kappa })^{\mathbf V} and \lambda-Borel function \mathbf B \in \mathbf V we have \nu = \mathbf B(\langle g'_{\rho(\varepsilon)}:\varepsilon < \lambda\rangle)
(b) if in \mathbf V[\mathbf G], ''_\gamma \in \prod\limits_{\zeta < \lambda} \theta_\zeta for \gamma < { \kappa } and the set \{(\gamma,\zeta):\gamma < { \kappa } and \zeta < \lambda and g''_\gamma(\zeta) \ne g'_\gamma(\zeta)\} has cardinality < \lambda then \bar g'' = \langle g''_\gamma:\gamma < { \kappa }\rangle is generic for {\mathbb P}'_{{ \kappa }} and \mathbf V[\bar g''] = \mathbf V[\bar g']; similarly for \mathbb{P} _ {\beta (*)},
(b)^+ similarly for any \gamma _ \bullet \le \kappa and \langle g''_ \gamma : \gamma < \gamma _ \bullet \rangle, really follows,
(c) \Vdash_{{\mathbb P}'_{{ \kappa }}} ``\let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3g\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}'_\gamma dominates (\prod\limits_{\varepsilon < \lambda} \theta_\varepsilon)^{\mathbf V[{\mathbb P}'_\gamma]}" for \gamma < { \kappa },
(d) if \langle \zeta(\gamma):\gamma < { \kappa }\rangle is an increasing sequence of ordinals < { \kappa } (from \mathbf V!), then \langle g'_{\zeta(\gamma)}:\gamma < { \kappa }\rangle is generic for {\mathbb P}'_{{ \kappa }} (over {\mathbf V});
(e) if \gamma \le { \kappa } then {\mathbb P}'_\gamma is (< \lambda)-strategically complete and satisfies the \lambda^+-c.c..
We shall use \boxplus_1 freely, this (mainly \boxplus _1(d)) had been the motivation for [Sh:1126].
To prove (*)_7 assume toward contradiction that this fails, and hence for some condition p^* \in {\mathbb P}'_{{ \kappa }} and {\mathbb P}'_{{ \kappa }}-name \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3f\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth} and \lambda-Borel function \mathbf B (from \mathbf{V}) and \rho_ \bullet \in {}^\lambda { \kappa } we have:
\circledast_0 p^* \Vdash_{{\mathbb P}'_{{ \kappa }}}
``\let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3f\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth} \in {}^\lambda \lambda and it does dominate \{ f^*_ \alpha : \alpha < \mu \}, equivalently ({}^\lambda \lambda)^{\mathbf V}" and \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3f\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth} = \mathbf B(\langle \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3g\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}'_{\rho_\bullet (i)}:i < \lambda\rangle).
Now let \chi be regular large enough and we choose \bar N = \langle N_\varepsilon:\varepsilon < \lambda \rangle such that:
\circledast_1
(a) N_\varepsilon is as in {\boxdot _{< \lambda,\theta _ *, \bar{\theta } } } for the forcing notion \mathbb{P} _{0, \beta (*)} (equivalently \mathbb{P} _{1, \beta (*)}, not \mathbb{P} '_\kappa), that is N_\varepsilon \in {\mathscr S}_{\chi, \mathbb{P} _{1, \beta (*) }} see {\boxdot _{< \lambda,\theta _ *, \bar{\theta } } } of 2.2(2),
(b)
\lambda_\varepsilon = \lambda (\varepsilon ) := {\rm otp}(N_\varepsilon \cap \lambda ) > \lambda^-_\varepsilon := \Sigma\{\|N_\zeta\|:\zeta < \varepsilon\} \ge \Sigma\{\lambda_\zeta:\zeta < \varepsilon\} \ge \varepsilon,
\bar N \restriction \varepsilon \in N_\varepsilon and{\rm otp } ( N_ \varepsilon \cap \kappa ) < \theta _ {\lambda ( \varepsilon )}, moreover {\rm otp}(\theta _* \cap N_ \varepsilon) < \theta _{\lambda (\varepsilon )}, (really \mathbb{P} _{1, \beta (*) } \in {\mathscr H} ( \theta _* )) hence \bigcup\limits_{\zeta < \varepsilon} N_\zeta \subseteq N_\varepsilon,
\lambda_{{\varepsilon}} is an inaccesible cardinal.
(c) \lambda, \kappa, \mu ,\theta _*, \bar\theta,\mathbf q, \bar{ \mathbb{P} }' , {\mathscr U}_*, p^*,\let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3f\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}, \mathbf B,\rho _ \bullet, \bar{ g } ' belong to N_\varepsilon.
Next choose f^* \in {}^\lambda \lambda, i.e. \in ({}^\lambda \lambda)^{\mathbf V}, such that:
\circledast_2 for arbitrarily large \varepsilon < \lambda for some \zeta \in [\lambda^-_\varepsilon,\lambda_\varepsilon) we have f^*(\zeta) > \lambda_\varepsilon, (we can demand more: for every large enough \varepsilon < \lambda).
For \varepsilon < \lambda let (\lambda_\varepsilon,\chi_\varepsilon,\mathbf j_\varepsilon, M_\varepsilon,{\mathbb A}_\varepsilon,\mathbf G ^+ _\varepsilon) be a witness for (N_\varepsilon,\mathbb{P} _{1, \beta (*)} , \chi) recalling \chi was chosen after \circledast_{1}(a) we have {\boxdot _{< \lambda,\theta _ *, \bar{\theta } } } from Definition 2.2(2) so \lambda_\varepsilon \in (\varepsilon,\lambda) is strongly inaccessible and \varepsilon < \zeta < \lambda \Rightarrow \lambda_\varepsilon < \lambda^-_\zeta < \lambda_\zeta, recalling \circledast_1 and noting \langle \lambda^-_{\varepsilon}:{\varepsilon} <\lambda\rangle is an increasing continuous sequence of cardinals below \lambda. Let \mathbf{G} ^ \dagger _ \varepsilon = \mathbf{G} ^+_ \varepsilon \cap \mathbf{j} _\varepsilon ( \mathbb{P} '_{\gamma ( * )}) recalling \mathbf{G} ^+_ {\varepsilon} \subseteq \mathbf{j} _ \varepsilon (\mathbb{P} _{1, \beta (*)}) and noting \mathbf{G} ^{\dagger } _ \varepsilon \subseteq \mathbf{j} _ \varepsilon (\mathbb{P} ' _{ \kappa }) \subseteq \mathbb{A} _ \varepsilon.
Let (for \varepsilon < \lambda):
\circledast_3
(a) v_\varepsilon = N_\varepsilon \cap { \kappa },
(b) {\kappa }_{\varepsilon} = {\kappa }(\varepsilon) = {\rm otp}(v_\varepsilon) and so {\kappa }_\varepsilon = \mathbf j_\varepsilon({ \kappa }) etc,
(c) \bar\gamma^\varepsilon =
\langle \gamma_i(\varepsilon):i < {\kappa }_\varepsilon \rangle list v_\varepsilon in increasing order,
(d) for i < {\rm otp}(v_\varepsilon), equivalently i < \mathbf j_\varepsilon({ \kappa }) = {\kappa }_ \varepsilon let \eta^\varepsilon_i = (\mathbf j_\varepsilon(\let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3g\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}'_{\gamma_i({\varepsilon})})) [\mathbf{G} ^ {\dagger } _ \varepsilon ] \in \prod\limits_{\zeta < \lambda_\varepsilon}\theta_\zeta \cap \mathbb{A} [\mathbf{G} ^{\dagger }_\varepsilon ] ,
(e) let \bar \eta^\varepsilon = \langle \eta^\varepsilon_i:i < {\kappa }_{\varepsilon}\rangle.
Note that clearly,
\circledast_4 for each \varepsilon < \lambda we have:
(a) \bar \eta^\varepsilon is generic for ({\mathbb A}_\varepsilon,\mathbf j_\varepsilon({\mathbb P}'_{{ \kappa }})), moreover
(b) if we change \eta^\varepsilon_i(\zeta) (legally, i.e. to an ordinal < \theta_\zeta) for < \lambda_\varepsilon pairs (i,\zeta) \in {\rm otp}(v_\varepsilon) \times \lambda_\varepsilon and get \bar \eta', then also \bar \eta' is generic for ({\mathbb A}_{\varepsilon},\mathbf j_\varepsilon ({\mathbb P}'_{{ \kappa }})), clearly N_ \varepsilon [\bar{ \eta } ^ \varepsilon ] = N_ \varepsilon [\bar{ \eta } ' ],
(c) there is a unique \mathbf{G}'_ \varepsilon a subset of \mathbb{P} _{1, \beta (*)} \cap N_ \varepsilon generic over N_ \varepsilon such that \mathbf{j} '' _ \varepsilon (\mathbf{G}' _ \varepsilon )= \mathbf{G} ^+_ \varepsilon so {\mathscr H} ( \chi _ \varepsilon ) = \mathbb{A} _\varepsilon [\mathbf{j} _ \varepsilon '' (\mathbf{G}'_ \varepsilon)]
[Why this equality holds? By \circledast_{1}(a) and the first line after \circledast_{2} i.e. 2.2(2); recall \mathbf{G} ^ \dagger_\varepsilon = \mathbf{G} ^+_ \varepsilon \cap \mathbf{j}_\varepsilon (\mathbb{P} ' _{{ \kappa } }) and we have (\mathbf{j} _ \varepsilon (\let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3 \crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth} _{\gamma _i(\varepsilon)}[\mathbf{G}' _\varepsilon ])= \eta ^ \varepsilon _i,]
(d) like \boxplus_1 with \mathbf V, \mathbb{P} _{1,\beta (*)}, \mathbf{G},\lambda there standing for {\mathbb A}_{\varepsilon},\mathbf j_\varepsilon(\mathbb{P} _{1,\beta (*)}) \mathbf{G}^+ _ \varepsilon ,\lambda_\varepsilon here.
Hence we have,
\circledast'_4 for \varepsilon < \lambda,
(a)
( \alpha ) let \Xi ^ \dagger _ \varepsilon : = \{ \bar{ \nu }: \bar{ \nu } = \langle \nu _i : i < \kappa _\varepsilon \rangle and for some {\mathbf G} ^\dagger \subseteq {\mathbb P}'_{ } \cap N_\varepsilon generic over N_ \varepsilon we have \nu _i \in \Pi _{ \xi < \lambda _ \varepsilon} \theta _ \xi satisfies \xi < \lambda _ \varepsilon \Rightarrow some \psi \in {\mathbf G}^\dagger - forces that for each i < \kappa _\varepsilon, \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3g \crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}' _{ \gamma _i (\varepsilon )} \restriction \xi = \nu _i \restriction \xi},
(\beta) for \bar{ \nu } \in \Xi ^\dagger _ \varepsilon let \mathbf{G} ^\dagger _ {\bar{ \nu }} be like \mathbf{G} ^\dagger above, (it is uniquely determined by \bar{ \nu }),
(\gamma ) let \Xi^+_\varepsilon = \{ \bar{ \nu } \in \Xi^\dagger_\varepsilon: there is a subset \mathbf{G} ^+ of \mathbf{j} _{ \varepsilon }(\mathbb{P} _{1, \beta (*)}) extending \mathbf{G}^\dagger _ {\bar{\nu}} and is generic over {\mathbb A}_{{\varepsilon}} ={\mathbf j}_{{\varepsilon}}''(N_{{\varepsilon}}) such that {\mathbb A}_{{\varepsilon}}[{\mathbf G}^{+}] = {\mathscr H}(\chi_{{\varepsilon}}) \} (so {\mathbf j}_{{\varepsilon}}^{-1}({\mathbf G}^{+}) is a subset of {\mathbb P}_{1, \beta(*)} \cap N_{{\varepsilon}} generic over N_{{\varepsilon}} and the Mostowski collapse of N_{{\varepsilon}}[{\mathbf j}_{{\varepsilon}}^{-1}({\mathbf G}^{+})] is {\mathscr H}(\chi_{{\varepsilon}})) .
(b)
(\alpha) Let \Xi ^ \bullet_ \varepsilon be the set of pairs ( \bar{ \nu }, \mathbf{G} ^+)= { (\bar{ \nu } , \mathbf{G} (+)) } such that: \bar{ \nu } \in \Xi ^{\dagger }_ \varepsilon and \mathbf{G} ^ + \subseteq \mathbf{j} _{ \varepsilon }(\mathbb{P} _{1, \beta (*)}) extending {\mathbf G}_{\overline{\nu}}^{\dagger} and is generic over {\mathbb A}_{{\varepsilon}}.
(\beta ) We may write (\bar{ \nu } , \mathbf{G} ^+_{\bar{ \nu } }) or just \bar{ \nu } though actually \bar{ \nu } does not determine \mathbf{G} ^+_{ \bar{ \nu} }; but of course \bar{ \nu } determines \mathbf{G} ^\dagger _{\bar{ \nu } }= \mathbf{G} ^+_{\bar{ \nu }} \cap \mathbf{j} _{ \varepsilon }(\mathbb{P} '_ { \gamma (*)}).
(\gamma ) Note that \bar{ \eta }^ \varepsilon belongs to \Xi^{\dagger } _ \varepsilon and even (\bar{ \eta } ^ \varepsilon , \mathbf{G} ) \in \Xi ^+_ \varepsilon when \mathbf{G} = \mathbf{G} ^+_\varepsilon.
[Why? See [Sh:1126, 3.28-3.32=Le53-Le67] and [Sh:1126, 4.27 = Le70]].
By the assumption toward contradiction, \circledast_0, and {\mathbb P}'_{{ \kappa }} being (< \lambda)-strategically complete, recalling \boxplus_1, there are \zeta(*),p^{**} and p^+ (recall p^* \in {\mathbb P}'_{{ \kappa }} \lessdot {\mathbb P}_{1,\beta(*)} is from \circledast_0 and \mathbb{P} _{0, \beta (*)} being dense in \mathbb{P} _{1, \beta (*)}) such that:
\circledast_5
(a) p^* \le p^{**} \in {\mathbb P}'_{{ \kappa }} and p^+ \in {\mathbb P}_{0,\beta(*)} satisfies {\mathbb P}_{1,\beta (*)} \models ``p^{**} \le p^+"; (we may add that \mathbb{P} '_{{ \kappa }} \models {``} p^{**}\le \phi " \Rightarrow \phi , p^+ are compatible in \mathbb{P} _{1, \beta (*)}),
(b) \zeta(*) < \lambda,
(c) p^{**} \Vdash_{{\mathbb P}'_{{ \kappa }}}
``f^*(\zeta) < \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3f\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}(\zeta) whenever \zeta(*) \le \zeta < \lambda" where f^* is from \circledast_2,
(d) if \gamma \in \text{ Dom}(p^+) then \eta^{p^+(\gamma)} is an object (not just a {\mathbb P}_{0,\gamma}-name) and has length \ge \zeta(*) (recall that \eta^{p^+(\gamma)} is the trunk of the condition p^+(\gamma), see clause (\alpha)(b) of Definition 1.4(1)).
Note that possibly Dom(p^+) \nsubseteq \cup\{v_\varepsilon:\varepsilon < \lambda\}. Choose \varepsilon (*) such that:
\circledast '_5 \varepsilon(*) < \lambda satisfies \lambda_{\varepsilon(*)} > \zeta(*) + |{\rm dom}(p^{+} )| and {\rm dom}(p^{+} ) \cap \cup \{ v_ \varepsilon : \varepsilon < \lambda \} \subseteq v_{\varepsilon (*)} and \gamma \in {\rm dom}(p^+) \Rightarrow \varepsilon(*) > \ell g(\eta^{p^+(\gamma)});
Recall clause (d) of \circledast_5 and |{\rm dom}(p^+)| < \lambda as p^+ \in {\mathbb P}_{0,\beta(*)} and {\mathbb P}_{0,\beta(*)} is the limit of a (< \lambda)-support iteration.
By \circledast_2 we can add (\exists \zeta)[\lambda^-_{\varepsilon(*)} \le \zeta < \lambda_{\varepsilon(*)} < f^*(\zeta)]. Our intention is to find q \in {\mathbb P}_{0,\beta(*)} above p^+ which (in {\mathbb P}_{1,\beta(*)}) is above some q' \in {\mathbb P}'_{{ \kappa }} which is ( N_ {\varepsilon (*)}, \mathbb{P} '_{ \kappa } )-generic, that is it forces \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3G\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}_{{\mathbb P}'_{{ \kappa }}} to include a generic subset of ({\mathbb P}'_{{ \kappa }})^{N_{\varepsilon(*)}} hence is induced by some \bar\nu as in \circledast'_4, recalling \circledast_4(b). Toward this in \circledast_6 below the intention is that p^+_{\kappa (\varepsilon (*))} will serve as q.
Let {\kappa }(*) = {\kappa }_{\varepsilon(*)} and \gamma_i for i <{\kappa }(*) be such that(This is used in \boxplus_3 and the proof of (*)_6. Not to be confused with \bar\gamma^{\varepsilon} of \circledast_3(c).) \langle \gamma_i:i < \kappa (*)\rangle list \{\beta^*_i:i \in v_{\varepsilon(*)}\} = {\mathscr U}_{**} = N_{\varepsilon ( * )} \cap {{\mathscr U} } _* in increasing order; recall {\mathscr U}_* = \{\beta^*_i:i < { \kappa }\} and i<j < { \kappa } \Rightarrow \beta^*_i < \beta^*_j and v_{\varepsilon(*)} \subseteq { \kappa } has order type {\kappa }(\varepsilon(*)) so \gamma _i = \gamma _i(\varepsilon (*)) from \circledast _3. Next let \gamma_{\kappa (*)} = {\kappa }(*) so \{\mathbf
j_{\varepsilon(*)}(\gamma):\gamma \in v_{\varepsilon(*)}\} = \kappa (*)
= \mathbf j_{\varepsilon(*)}({ \kappa }). Recall that { \kappa } = {\rm cf}(\kappa) > \lambda,{\rm otp}(v_{\varepsilon(*)}) = {\rm otp}(N_{\varepsilon(*)} \cap { \kappa }) hence N_{\varepsilon(*)} \models `` { \kappa } (*)= \kappa _{\varepsilon ( *)} is a regular cardinal > \lambda_{\varepsilon(*)}"; hence:
(*) {\kappa }(*) is really a regular cardinal so call it \sigma.
Now we define a game \Game as follows(The idea is to scatter the \eta^{\varepsilon(*)}_{\gamma_i}’s. Why not use the original places? as then we shall have a problem in \circledast_6; the scattering is helpful because we are relying on 1.10 and 1.11.):
\boxplus_2
(A) each play lasts {\kappa }(*)+1 = \sigma +1 moves and in the i-th move:
(a) if i=j+1 the antagonist player chooses \xi_j = \xi(j) < \sigma such that j_1 < j \Rightarrow \zeta (j_1) < \xi(j),
(b) then, if i=j+1 the protagonist chooses \zeta_j = \zeta(j) \in (\xi(j),\sigma), but there are more restrictions implicit in \boxplus_3 below,
(c) in any case (that is, also in the cases i \le \sigma is a limit ordinal or zero) the protagonist also chooses p^+_i,\bar\nu^i such that \boxplus_3 below holds.
(B) in the end of the play the protagonist wins the play iff he always has a legal move and in the end:
(a) p^+_ \sigma is ( {\mathbb P} '_{\gamma ( * ) }, N_{\varepsilon (*)} )-generic, note the condition is not a member of the same forcing, so we mean that p^+_ \sigma forces (for \mathbb{P} _{1, \beta (*)}) that the intersection of the generic with \mathbb{P} '_{} \cap N_{\varepsilon (*) } is generic over N_{\varepsilon (*)},
(b) \{\zeta(i):i < \sigma \} \in {\mathbb A}_{\varepsilon(*)}; note that trivially it belongs to M_{\varepsilon(*)} = {\mathbb A}_{\varepsilon(*)}[\mathbf{G}^+ _ \varepsilon ] {\mathscr H} ( \chi _ \varepsilon ), see \circledast_4(c).
(c) note that we do not demand that \bar{ \nu } ' = \langle \nu _{\gamma _i}: i < \sigma \rangle belongs to \Xi ^+_ \varepsilon, we demand only that it belongs to \Xi ^\dagger_ \varepsilon and even \Xi_{{\varepsilon}}'.
where,
\boxplus_3
(a) p^+_i \in {\mathbb P}_{0,\gamma_i},
(b) if j<i then {\mathbb P}_{0,\gamma_i} \models ``p^+_j \le p^+_i",
(c) if \gamma \in \cup\{\text{Dom}(p^+_j):j<i\} then p^+_i \restriction \gamma\Vdash_{{\mathbb P}_{0,\gamma_i}} ``\let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}^{p^+_i(\gamma)} has length \ge i and \ge \lambda_{\varepsilon(*)}" moreover \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}^{p^+_i(\gamma)} is an object, \eta^{p^+_i(\gamma)},
(d) {\mathbb P}_{0,\gamma_i} \models ``p^{+} \restriction \gamma_i \le p^+_i", (p^+ is from \circledast _5(a)),
(e)
(\alpha) \bar \nu^i = \langle \nu_{\gamma_j}:j <
i\rangle and \nu_{\gamma_j} \in
\prod\limits_{\iota < \lambda_{\varepsilon(*)}} \theta_\iota,
(\beta) {\mathbf G}_{{\varepsilon}(*), i}^{\dagger \dagger} is a subset of {\mathbb P}_{\gamma_{i}} \cap N_{{\varepsilon}(*)} generic over N_{{\varepsilon}(*)},
(\gamma) {\mathbf G}_{{\varepsilon}(\ast), i}^{\dagger} = {\mathbf j}_{{\varepsilon}(*)}''({\mathbf G}_{{\varepsilon}(*), i}^{\dagger \dagger}) so is a subset of {\mathbf j}_{{\varepsilon}}''({\mathbb P}_{\gamma_{2}}') generic over {\mathbb A}_{{\varepsilon}(*)},
(\delta) \nu_{j} = \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}_{\gamma_{j}}[{\mathbf G}_{{\varepsilon}(*), i}^{\dagger \dagger}] for j < i.
(f) for j<i we have \nu_{\gamma_j} \trianglelefteq \eta^{p^+_i(\gamma_j)} so p^+_i \restriction \gamma_j \Vdash ``\nu_{\gamma_j} \triangleleft \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3g\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}'_{\gamma_j}" recalling \boxplus_1,
(g) for j<i (recall \bar\eta^\varepsilon is from \circledast_3(d)) we have ( \alpha ) or ( \beta ), where:
(\alpha) \nu_{\gamma_j} =
\eta^{\varepsilon(*)}_{\gamma_{\zeta(j)}} recalling \eta^{\varepsilon(*)}_{\gamma_j} is from \circledast_3(d),
(\beta) \gamma_j \in \text{ Dom}(p^{+}) and \{\iota < \lambda_{\varepsilon(*)}:\eta^{\varepsilon(*)}_{\zeta(j)}(\iota) \ne \nu_{\gamma_j}(\iota)\} is a bounded subset of \lambda_{\varepsilon(*)}.
(h) p^+_i is an upper bound of {\mathbf G}_{{\varepsilon}(*), i}^{\dagger \dagger} hence is a (N_ {\varepsilon(\ast)}, \mathbb{P}_{\gamma_{i}}')-generic in the natural sense, (actually follows from clause (g), see later in \odot in the beginning of the proof of 2.8).
We shall prove,
\circledast_6 in the game \Game:
(a) the antagonist has no winning strategy,
(b) at stage i, if \langle \zeta ( j ) : j < i \rangle \in \mathbb{A} _ \varepsilon then the protagonist has a legal move, moreover for any \zeta(i) \in (\xi(i),\sigma) large enough the protagonist can choose it.
Why \circledast_6 suffice?
By clause (a) of \circledast_6 we can choose a play \langle(\xi(i),\zeta(i),p^+_i,\bar\nu^i, {\mathbf G}^{\dagger \dagger}_{{\varepsilon}(*), \sigma}, {\mathbf G}_{{\varepsilon}(*), \nu}^{\dagger}) :i \le \sigma\rangle in which the protagonist wins. Recalling {\mathbb P}'_{{ \kappa }} \lessdot {\mathbb P}_{1,\beta(*)} and {\mathbb P}_{0,\beta(*)} is a dense sub-forcing of {\mathbb P}_{1,\beta(*)}, clearly,
\circledast_7 there is p such that:
(a) p \in {\mathbb P}'_{{ \kappa }},
(b) if {\mathbb P}'_{{ \kappa }} \models ``p \le p'" hence p' \in {\mathbb P}'_{ \gamma (*)} then p',p^+_\sigma are compatible in {\mathbb P}_{1,\beta(*)},
(c) p is above p^{**} and it forces that \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3g\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}'_{\gamma_i} {\restriction} \lambda_{\varepsilon(*)} = \nu_{\gamma_ {\zeta (i) }} for i < \sigma and \mathbf{j} _{\varepsilon (*)}( \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3G\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth} _{ {\mathbb P}' _{ \beta ( * ) } } \cap N _ {\varepsilon (*)}) = {\mathbf G}^ {\dagger } _{\langle \nu _{\gamma _i }: i < \sigma \rangle} \in \Xi ^ {\dagger }_{\varepsilon (*) }.
Then on the one hand,
\circledast_{7.1} p \in {\mathbb P}'_{{ \kappa }} being above p^{**} forces f^* \restriction [\zeta(*),\lambda) < \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3f\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth} \restriction
[\zeta(*),\lambda) hence f^* \restriction
[\zeta(*),\lambda_{\varepsilon(*)}) < \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3f\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}
\restriction [\zeta(*),\lambda_{\varepsilon(*)}) recalling that \zeta(*) < \lambda_{\varepsilon(*)}, see \circledast_5 and the choice of {\varepsilon}(*) immediately after \circledast_5.
On the other hand,
\circledast_{7.2} {\mathbf G}_{{\varepsilon}(*), \sigma}^{\dagger \dagger} is a subset of {\mathbb P}' \cap N_{{\varepsilon}} generic over N_{{\varepsilon}}.
[Why? By \circledast_{2}(e)(\beta) and the choice of the play.]
\circledast_{7.3} p_{\sigma}^{+} is an upper bound of {\mathbf G}_{{\varepsilon}(*), \nu}^{\dagger \dagger}.
[Why? By \circledast_{2}(h) and the choice of the play.]
\circledast_{7.4} p is an upper bound of {\mathbf G}_{{\varepsilon}(*), \nu}^{\dagger \dagger} in {\mathbb P}.
[Why? By \circledast_{7}(b), \circledast_{7.1} and \circledast_{7.2}].
\circledast_{7.5} p is (N_{{\varepsilon}(*), \sigma}, {\mathbb P}')-generic.
[Why? If p_{i}^{+} is not an upper bound of {\mathbf G}_{{\varepsilon}(*), i}^{\dagger \dagger} (in {\mathbb P}_{1, \gamma_{i}}) then there are p_{i}'' and r \in G_{{\varepsilon}(*), i}^{\dagger \dagger} such that {\mathbb P}_{1, \gamma_{i}} \models“p_{i}^{+} \leq p'” and p', r are compatible in {\mathbb P}_{1, \gamma_{i}}. As {\mathbb P}_{0, \gamma_{i}} is a dense subset of {\mathbb P}_{1, \gamma_{i}} and {\mathbb P}_{\gamma_{i}}' \cap N_{{\varepsilon}(*)} is a subset of {\mathbb P}_{1, \gamma_{i}} of cardinality \leq \Vert N_{{\varepsilon}(*)} \Vert < \lambda_{{\varepsilon}}, there is p'' \in {\mathbb P}_{0, \gamma_{i}} above p_{i}' which decide the value of {\mathbb P}_{\gamma_{i}}' \cap N_{{\varepsilon}(*)}].
As \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3f\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth} \in N_{\varepsilon(*)} it follows from \circledast_{7.5} that:
\circledast_{7.6} p \Vdash ``\let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3f\crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth} \restriction \lambda_{\varepsilon(*)} is a function from \lambda_{\varepsilon(*)} to \lambda_{\varepsilon(*)}".
Together \circledast_{7.1} and \circledast_{7.6} give a contradiction by the choice of f^* in \circledast_2 and of \varepsilon(*) above which implies that \let\theaccent\tilde\skewfactor-
\mathpalette\oalign{3 f \crcr\hidewidth
\vbox to.2ex{\hbox{3}\vss}\hidewidth}( \zeta ) > f^*(\zeta ) > \lambda _{\varepsilon (*)} for some \zeta < \lambda _{\varepsilon (*)} hence \circledast_6 is enough. In Lemma 2.8 below we show that \circledast_6 is true; so we are done.