In this work we study complexity classes in monotone computation. Our main contributions are the following: ffl A consistent framework for monotone computation, including monotone analogues of many standard computational models. We define monotone simulati

s oht eafmliia nroiotnsof modl iendpeenencedan d unviresa plrobemsl acrry trough. hO nhte ohert had thnre eermia snmoef unadmental d ereincse ebwteenmon tooenan d gneraelc molexitp,ya n padr to our goafl si tond nad xplein asuh ci erdecn

es. Erly raseercharesin thea reaof lowerbo udn fsrom notonoec icrits uewermot iatedv byth efo lowlign ohe:pperha pst he emhtdosd veleopd eo dtae wilth onmtoon coepumattionw uolde evtnalul by eaaptdbla teoh ande glenrae lomcupttaoi ans wlel La.te rerustl 3s2,4] h3vea shownt ishh pe to oe naibv.e We sek to cleraiyfw hat( ifan ) ryeevlacen mnoooten compleixy ths toa gnerae lcmpoleixyt .We rgaeutha tonmtono seperaatons aireint erstein whgen heyt eltlus s oetmhing idffreet nhatnwha t ew nkw oi nhet geenar mlodl.eIn tehse siutatonsi wen dtat there hsiso eminh erntlye onnmnotonoe isulatmon ioing gn ion th geneealrsett ing, an denhc ethis simlutaoin siin a|p rcies eessenstr|ongre hanttho s simelutions ahicwha re kownn to b moentone.oTh t ais,m ontono eseapartois mny aopit ontut h nenoboiovsu sretgthsn o fegneal comprtautoi.nFor xample,e hetfo llowngit ehcnquis eare niteestirng i that tnhye reaprovab yl nonmnotooen I:mermma nna Sdezepclsnyi's einducive tcuntingo ethnciqeu (haptecr3),uiversaln rtaversl saeuqecnes (chpter 3a,) Brrinagon'tsc ommtuaito tenhniqcue( haptcre )4. nI hist hacpetr e wgvei nainform alove rviwe nomo ntono coemlpeixt ynda mtovaie the pretset norkw I.n capthre2 we g ie a movr pereisc efamreowrkfor omnoton coemptautoi nand hsw othee qivuleaneco fv riaosum odesl .nI chatep 3 wer hsw otah ntonetedrminsitc iogsplcaei nots closed udner coplmemntateoini nt eh onotmoen rafmeorwk, ulnki Iemermanms s'iumatlin in oth easecof gneealr ocmptatiun. Inoc hapetr 4w preseen prteilmianryw or towardk asc onjetcre uotmiatevd y theb preiovsur sulte th;e cnoejcurt esi htt aaBrringtnos's imluaiotn alo hassn oonmtone anoaloueg .I ncapthre we 5hswo hat the mtootnoen analogeu o logfsace cpmoptautinoi sst rctlyimor eowerpfu lthan mnootoe nlo-geptdhc icrutsi Fi.nlal yi nhapter 6c wep reens tsveerla rmaeninigpr boemlss uggsetdeb y thi wosk.r

12