Rules not feeding each other in ace transfer

ACE
#1

I have a set of transfer rules that each work individually but don’t seem to be feeding each other as I expect them to. I’m using ace for transfer. Here is an example pair of rules:

be-located-mtr := monotonic_mtr &                                                                                       
[ INPUT [ RELS < [ PRED "_be_located_v_rel",                                                                            
  LBL #lbl,                                                                                                             
  ARG0 #arg0,                                                                                                           
  ARG1 #arg1,                                                                                                           
  ARG2 #arg2 ] >,                                                                                                       
      HCONS < > ],                                                                                                      
  OUTPUT [ RELS < [ PRED "_in_p_rel",                                                                                   
   LBL #lbl,                                                                                                            
   ARG0 #arg0,                                                                                                          
   ARG1 #arg1,                                                                                                          
   ARG2 #arg2 ] > ]].     

in-be_in-mtr := monotonic_mtr &                                                                                         
[ INPUT [ RELS < [ PRED "_in_p_rel",                                                                                    
                   LBL #lbl,                                                                                            
                   ARG0 #arg0,                                                                                          
                   ARG1 #arg1,                                                                                          
                   ARG2 #arg2 ] >],                                                                                     
  OUTPUT [ RELS < [ PRED "_be.in_v_rel",                                                                                
                    LBL #lbl,                                                                                           
                    ARG0 #arg0,                                                                                         
                    ARG1 #arg1,                                                                                         
                    ARG2 #arg2 ] > ]].

And the type monotonic_mtr and its supertype are defined like this:

monotonic_mtr := mrs_transfer_rule &                                                                                    
[ CONTEXT [ LTOP #h, INDEX #i ],                                                                                        
  INPUT [ LTOP #h, INDEX #i ],                                                                                          
  OUTPUT [ LTOP #h, INDEX #i ] ].        


mrs_transfer_rule :=  top &                                                                                             
[ FILTER mrs,                                                                                                           
  CONTEXT mrs,                                                                                                          
  INPUT mrs,                                                                                                            
  OUTPUT mrs,                                                                                                           
  FLAGS flags ].

Do I have the wrong expectation about rules feeding each other?

(For anyone wondering why I’d want that pair of rules, this is in the context of the MT demo at the end of Ling 567, accommodating slightly different approaches to this construction in different grammars.)

#2

p.s. I meant to tag @sweaglesw here :slight_smile: