Subject: Res: Do phonemes = sounds? From: Daniel Silva <silvadmr(at)BOL.COM.BR> Date: Mon, 13 Jun 2005 14:06:27 -0300--------------Boundary-00=_RU81QL80000000000000 Content-Type: Multipart/Alternative; boundary="------------Boundary-00=_RU81LVC0000000000000" --------------Boundary-00=_RU81LVC0000000000000 Content-Type: Text/Plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable We have in my language (Brazilian Portuguese) an interesting example concerning the question (especially the post by Dr. Port). Generally, our vowel system is described as containing seven vowel phonemes, the contras= t between + ATR and =96ATR mid vowels being functional only in the stressed syllables. =0D However, in spite of the few minimal pairs for the -/+ ATR contrast, we h= ave words (many of them high frequency words) in witch we can employ both the= =96 ATR and the + ATR vowel in the stressed syllable (examples: cr[o]sta/cr[O]sta, p[o]=E7a/p[O]=E7a, [e]xtra/[E]xtra=85). Additionally, = the employment of these vowels shows some predictability in certain cases (bu= t we have always the counterexamples). I=92m less then a novice, but I thin= k patterns like this challenge the notion of =93phoneme=94.=0D =0D =0D -------Mensagem original-------=0D =0D De: Robert Port=0D Data: 06/07/05 15:58:56=0D Para: AUDITORY(at)LISTS.MCGILL.CA=0D Assunto: Do phonemes =3D sounds?=0D =0D I can't resist responding to this issue. I have spent my career trying=0D to figure out what words are made of. I have finally come to the=0D conclusion that:=0D =0D BOTH PHONEMES AND PHONES ARE INTUITIVELY PERSUASIVE PRIMARILY=0D BECAUSE OF OUR LIFELONG EXPERIENCE WITH ALPHABETS.=0D =0D Speech sounds are very short (15-20 per second), and the relevant=0D motor gestures are mostly invisible (tongue, glottis, velum, etc). And=0D those of us in the European cultural tradition learn to use letters=0D beginning as young as 2. Letters are a great engineering solution to=0D preserving language in graphic form and our education system assures=0D that we all become proficient at thinking about speech in letter-like=0D terms from an early age. But phones and phonemes inherit many=0D graphic properties from letters:=0D * SERIAL ORDER (no temporal patterns allowed),=0D * NONOVERLAPPING (hence the artificial `coarticulation problem'),= =0D * PERFECTLY CONTRASTIVE FROM ONE ANOTHER (no near contrasts or=0D partial contrasts)=0D * STATIC (diphthongs, glides and affricates present awkward=0D inconsistencies).=0D =0D In the late 19th C, de Courtenay proposed the notion of the `phoneme'=0D which was very quickly adopted by phoneticians and linguists and=0D treated as a great discovery about human language. In fact, all that=0D happened is that scientists began to think seriously about the=0D psychological representation of language (which had been largely=0D ignored earlier) and thought `Maybe we have something in our heads=0D that represents words the way letters do. There must be something=0D analogous to letters to keep words distinct from each other. There=0D must be PHONEMES!' The phoneme was not discovered, it was just=0D postulated by analogy with alphabetical written language.=0D =0D So how ARE words `represented' in the head?: By gestures and gesture=0D components of various sizes - from feature size to syllable size to=0D the size of whole phrases - WHATEVER STATISTICAL REGULARITIES SPEAKERS=0D HAPPEN TO PICK UP as they learn how TO talk. Each language has its=0D own conventional solution to keeping utterances distinct. Of course,=0D languages have a phonological system, something a little like an=0D `inventory' of possible sound contrasts in various positions. This=0D system should be thought of as a social institution that children=0D learn to adapt their speech habits to. But these inventories are very=0D different from a cognitive spelling system. Eg, they always have many=0D uncertainties, places where you cannot tell which phonemes (letters)=0D to employ: what is the V in `beer'? (same as bead or same as bid?),=0D what is the stop in `stow'? Same as tow or same as doe? What is the=0D pstop butter? Same a butt or same as Bud? Or neither? (But the data=0D show two slightly different flaps are used in ladder and latter!) If=0D language were REALLY spelled with a phonological alphabet, then these=0D uncertainties could not happen.=0D =0D I have a couple mspts on my website addressing these=0D issues. Obviously there are a great many empirical implications of=0D this iconoclastic hypothesis: implications about how speech can be=0D recognized (by humans or machine), for how languages are learned, how=0D they change over time, differences between the linguistic intuitions=0D of alphabet-literates vs. nonalphabet-literates, constraints on=0D phonological language games (like pig-latin), etc etc. Basically ALL=0D the data come out just as this hypothesis predicts. I claim, in fact:=0D =0D NO DATA WHATEVER, (aside from the powerful intuitions of lay=0D people and linguists educated in the alphabetic tradition) SUPPORT A=0D SEGMENTAL (that is, C and V) DESCRIPTION OF LANGUAGE!=0D =0D I would like to hear the evidence if anyone wishes to challenge this=0D claim.=0D =0D I have a couple papers on my website presenting these=0D arguments in more detail.=0D http://www.cs.indiana.edu/~port/pubs.html=0D =0D Bob Port=0D =0D ( ( ( O ) ) ) ( ( ( O ) ) ) ( ( ( O ) ) )=0D Lingstcs/Comp Sci/Cogntv Sci=0D ROBERT F. PORT 330 Memorial Hall, Indiana University=0D Bloomington, Indiana 47405=0D 812-855-9217 Fx 812-855-5363=0D http://www.cs.indiana.edu/~port --------------Boundary-00=_RU81LVC0000000000000 Content-Type: Text/HTML; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable <HTML><HEAD> <META http-equiv=3DContent-Type content=3D"text/html; charset=3Diso-8859-= 1"> <META content=3D"IncrediMail 1.0" name=3DGENERATOR></HEAD> <BODY style=3D"BACKGROUND-POSITION: left 0px; FONT-SIZE: 12pt; MARGIN: 0p= x 10px 10px; COLOR: #00005b; BACKGROUND-REPEAT: repeat; FONT-FAMILY: Aria= l" text=3D#00005b bgColor=3D#eff3f7 background=3Dcid:33AA58F6-CAD1-46A9-B= 334-261C354B10DC scroll=3Dyes SIGCOLOR=3D"0" ORGYPOS=3D"0"> <TABLE id=3DINCREDIMAINTABLE cellSpacing=3D0 cellPadding=3D2 width=3D"100= %" border=3D0> <TBODY> <TR> <TD id=3DINCREDITEXTREGION style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 0px= ; FONT-SIZE: 12pt; PADDING-BOTTOM: 0px; CURSOR: auto; PADDING-TOP: 0px" v= Align=3Dtop width=3D"100%"> <DIV> <P class=3DMsoNormal style=3D"MARGIN: 0cm 0cm 0pt"><SPAN lang=3DEN-US sty= le=3D"mso-ansi-language: EN-US"><FONT color=3D#000000><FONT face=3D"Times= New Roman">We have in my language (Brazilian Portuguese) an interesting = example concerning the question (especially the post by Dr. Port). Genera= lly, our vowel system is described as containing seven vowel phonemes, th= e contrast between + ATR and =96ATR mid vowels being functional only in t= he stressed syllables. <?xml:namespace prefix =3D o ns =3D "urn:schemas-m= icrosoft-com:office:office" /><o:p></o:p></FONT></FONT></SPAN></P> <P class=3DMsoNormal style=3D"MARGIN: 0cm 0cm 0pt"><SPAN lang=3DEN-US sty= le=3D"mso-ansi-language: EN-US"><FONT color=3D#000000><FONT face=3D"Times= New Roman">However, in spite of the few minimal pairs for the -/+ ATR co= ntrast, we have words (many of them high frequency words) in witch we can= employ both the =96 ATR and the + ATR vowel in the stressed syllable (ex= amples: cr[o]sta/cr[O]sta, p[o]=E7a/p[O]=E7a, [e]xtra/[E]xtra=85). Additi= onally, the employment of these vowels shows some predictability in certa= in cases (but we have always the counterexamples). I=92m less then a novi= ce, but I think patterns like this challenge the notion of =93phoneme=94.= <o:p></o:p></FONT></FONT></SPAN></P> </DIV> <DIV> </DIV> <DIV id=3DIncrediOriginalMessage><I>-------Mensagem original-------</I></= DIV> <DIV> </DIV> <DIV id=3Dreceivestrings> <DIV dir=3Dltr style=3D"FONT-SIZE: 11pt"><I><B>De:</B></I> <A href=3D"mai= lto:port(at)CS.INDIANA.EDU">Robert Port</A></DIV> <DIV dir=3Dltr style=3D"FONT-SIZE: 11pt"><I><B>Data:</B></I> 06/07/05 15:= 58:56</DIV> <DIV dir=3Dltr style=3D"FONT-SIZE: 11pt"><I><B>Para:</B></I> <A href=3D"m= ailto:AUDITORY(at)LISTS.MCGILL.CA">AUDITORY(at)LISTS.MCGILL.CA</A></DIV> <DIV dir=3Dltr style=3D"FONT-SIZE: 11pt"><I><B>Assunto:</B></I> Do phonem= es =3D sounds?</DIV></DIV> <DIV> </DIV> <DIV>I can't resist responding to this issue. I have spent my career tryi= ng</DIV> <DIV>to figure out what words are made of. I have finally come to the</DI= V> <DIV>conclusion that:</DIV> <DIV> </DIV> <DIV> BOTH PHONEMES AND PH= ONES ARE INTUITIVELY PERSUASIVE PRIMARILY</DIV> <DIV>BECAUSE OF OUR LIFELONG EXPERIENCE WITH ALPHABETS.</DIV> <DIV> </DIV> <DIV>Speech sounds are very short (15-20 per second), and the relevant</D= IV> <DIV>motor gestures are mostly invisible (tongue, glottis, velum, etc). A= nd</DIV> <DIV>those of us in the European cultural tradition learn to use letters<= /DIV> <DIV>beginning as young as 2. Letters are a great engineering solution to= </DIV> <DIV>preserving language in graphic form and our education system assures= </DIV> <DIV>that we all become proficient at thinking about speech in letter-lik= e</DIV> <DIV>terms from an early age. But phones and phonemes inherit many</DIV> <DIV>graphic properties from letters:</DIV> <DIV> * SERIAL ORDER (no t= emporal patterns allowed),</DIV> <DIV> * NONOVERLAPPING (he= nce the artificial `coarticulation problem'),</DIV> <DIV> * PERFECTLY CONTRAST= IVE FROM ONE ANOTHER (no near contrasts or</DIV> <DIV>partial contrasts)</DIV> <DIV> * STATIC (diphthongs= , glides and affricates present awkward</DIV> <DIV>inconsistencies).</DIV> <DIV> </DIV> <DIV>In the late 19th C, de Courtenay proposed the notion of the `phoneme= '</DIV> <DIV>which was very quickly adopted by phoneticians and linguists and</DI= V> <DIV>treated as a great discovery about human language. In fact, all that= </DIV> <DIV>happened is that scientists began to think seriously about the</DIV> <DIV>psychological representation of language (which had been largely</DI= V> <DIV>ignored earlier) and thought `Maybe we have something in our heads</= DIV> <DIV>that represents words the way letters do. There must be something</D= IV> <DIV>analogous to letters to keep words distinct from each other. There</= DIV> <DIV>must be PHONEMES!' The phoneme was not discovered, it was= just</DIV> <DIV>postulated by analogy with alphabetical written language.</DIV> <DIV> </DIV> <DIV>So how ARE words `represented' in the head?: By gestures and gesture= </DIV> <DIV>components of various sizes - from feature size to syllable size to<= /DIV> <DIV>the size of whole phrases - WHATEVER STATISTICAL REGULARITIES SPEAKE= RS</DIV> <DIV>HAPPEN TO PICK UP as they learn how TO talk. Each languag= e has its</DIV> <DIV>own conventional solution to keeping utterances distinct.  = ;Of course,</DIV> <DIV>languages have a phonological system, something a little like an</DI= V> <DIV>`inventory' of possible sound contrasts in various positions. This</= DIV> <DIV>system should be thought of as a social institution that children</D= IV> <DIV>learn to adapt their speech habits to. But these inventories are ver= y</DIV> <DIV>different from a cognitive spelling system. Eg, they alwa= ys have many</DIV> <DIV>uncertainties, places where you cannot tell which phonemes (letters)= </DIV> <DIV>to employ: what is the V in `beer'? (same as bead or same as bid?),<= /DIV> <DIV>what is the stop in `stow'? Same as tow or same as doe?&n= bsp; What is the</DIV> <DIV>pstop butter? Same a butt or same as Bud? Or neither? (Bu= t the data</DIV> <DIV>show two slightly different flaps are used in ladder and latter!)&nb= sp; If</DIV> <DIV>language were REALLY spelled with a phonological alphabet, then thes= e</DIV> <DIV>uncertainties could not happen.</DIV> <DIV> </DIV> <DIV> I have a couple mspt= s on my website addressing these</DIV> <DIV>issues. Obviously there are a great many empirical implications of</= DIV> <DIV>this iconoclastic hypothesis: implications about how speech can be</= DIV> <DIV>recognized (by humans or machine), for how languages are learned, ho= w</DIV> <DIV>they change over time, differences between the linguistic intuitions= </DIV> <DIV>of alphabet-literates vs. nonalphabet-literates, constraints on</DIV= > <DIV>phonological language games (like pig-latin), etc etc. Basically ALL= </DIV> <DIV>the data come out just as this hypothesis predicts. I cla= im, in fact:</DIV> <DIV> </DIV> <DIV> NO DATA WHATEVER, (a= side from the powerful intuitions of lay</DIV> <DIV>people and linguists educated in the alphabetic tradition) SUPPORT A= </DIV> <DIV>SEGMENTAL (that is, C and V) DESCRIPTION OF LANGUAGE!</DIV> <DIV> </DIV> <DIV>I would like to hear the evidence if anyone wishes to challenge this= </DIV> <DIV>claim.</DIV> <DIV> </DIV> <DIV> I have a couple pape= rs on my website presenting these</DIV> <DIV>arguments in more detail.</DIV> <DIV> <A href=3D"http://www.cs.indiana.edu/~port/pubs.html">h= ttp://www.cs.indiana.edu/~port/pubs.html</A></DIV> <DIV> </DIV> <DIV> Bob Port</DIV> <DIV> </DIV> <DIV> ( ( ( O )&nb= sp; ) ) ( ( (  = ;O ) ) ) ( ( = ( O ) ) )</DIV> <DIV> &n= bsp; Lingstcs= /Comp Sci/Cogntv Sci</DIV> <DIV> ROBERT F. PORT 330 M= emorial Hall, Indiana University</DIV> <DIV> &n= bsp; Blooming= ton, Indiana 47405</DIV> <DIV> 812-855-9217 &= nbsp;Fx 812-855-5363</DIV> <DIV> <A href=3D"http://www.cs.indiana.edu/~port"= >http://www.cs.indiana.edu/~port</A></DIV></TD></TR> <TR> <TD id=3DINCREDIFOOTER width=3D"100%"> <TABLE cellSpacing=3D0 cellPadding=3D0 width=3D"100%"> <TBODY> <TR> <TD width=3D"100%"></TD> <TD id=3DINCREDISOUND vAlign=3Dbottom align=3Dmiddle></TD> <TD id=3DINCREDIANIM vAlign=3Dbottom align=3Dmiddle></TD></TR></TBODY></T= ABLE></TD></TR></TBODY></TABLE><SPAN id=3DIncrediStamp><SPAN ltr??><A hre= f=3D"http://www.incredimail.com/index.asp?id=3D422&lang=3D22"><IMG al= t=3D"" hspace=3D0 src=3D"cid:8170D9BF-0C1A-47C2-9AE8-6E9EF8F9A95D" align=3D= baseline border=3D0></A></SPAN></SPAN></BODY></HTML> --------------Boundary-00=_RU81LVC0000000000000-- --------------Boundary-00=_RU81QL80000000000000 Content-Type: image/gif; name="IMSTP.gif" Content-Transfer-Encoding: base64 Content-ID: <8170D9BF-0C1A-47C2-9AE8-6E9EF8F9A95D> R0lGODlhDAJMAMQAAHYdAP/yACLVxP/7nAcA//AHy/7SAPy1AJ5IDPWYAP/93aA0Ad9xAfwMDP/5 eP///7Crm9rQvfejmv3Bav+D7ebh1K1zJv+uNP/6Mc2XAAqDg/tvaNewAL2NPgAAAAAAACH/C05F VFNDQVBFMi4wAwEAAAAh+QQJHgAfACwAAAAADAJMAAAF/+AnjmRpnmiqrmzrvnAsz3Rt33iu73zv /8CgcEgsGo/IpHLJbL480Kh0Sq1ar9isdsvter/gsHhMLpvP6LR6zW6XnfC4fE6v2+/4vH7P7/v/ gIGCg4SFhoeIiYqLjI2Oj5CRkpOUlZaXmJmam5ydnp+goaKjpKWmp6ipqqusMVCtsCWvH7M5BAQi t3e6sX5UNrW1Q1MjvyRRxVW0HsfMy69WycjN06yzwrLOJ7e8H7zddN+4vXnYNeZEws7B2s8m6u3s 2fMi8sXk6PXtJN3j4Hv/yNExB21dvGnE2CnURkzavmUo7EHEBq9ZMnr66OVzeFDetWrZFl6EqA+d lBQFM/86HCmSZImA3vyNi8ktl8wR3Ga+rIkzp82fMYEKtEOwILOPKu0FI4mU6cOR7xhKdamyqkuK BzVudBq1JNauFrmKPdE06tGUWpPuQwdTHFC3QePC7Pn2ply7Q+sYa8rXIMu1fsVuzGcs41eoVwGH hdowrdfAHMH+VWvV8UnBUhEGPrxzm124oPHy83m3rum8eh/2pZy48urETz8i/cqZqkSrtUuCfV2R qjSOmzOTbWw5OGt3lYWOPh3aJunlpaM3Rz1QtXHMk8O+Roz7+rPL3RfXzp2cN+Sxu627Vr/CPGvy nU1Mn69T/uf7p6kTZe++9VXsvpk0lX+39VZVbhJlVdz/ceg5RguDvv033HvXHZZPP8zhF5039mXo YVz6xbGXd+C509J32ZUYkmYWFQgYeCUWBdJ3scV4GVYzorhSZDqa9RSFPPZ120794JXTTEf2NFeS zhXJE1whRjnJVlK6MJcOV1appSVUCnTkl2CGKeaYZJYZ5pZopqnmmmy26eabcMYp55x01mnnnXjm qeeefPbp55+ABirooIQWauihiCaq6KKMNuroo5BGKmkdAhyQAAwCVPCBABrY0EADKhQg6qgUlFqq qKZSoAICJBhgAACwArDArAwwwcCsC8Qa6wiw5nXrrLr2+gMAtYrwa67BGgvApDYkcMAJmYrg7AgC CLCp/7UjBBDAB9NC26kIn56gwbiqFvABqg1QkG4BpbaAwLuuupqAAlAA4EEEDtyKxAIMDKDAvwD/ e+uyHzAgLAoILMDqCwQX/C4CwTZsAwMPv6urDwZ/wK+/AQM8cA8Nb9yxAhZ08LAIEjMLg6vQYhvA pSSIWoK21T7bsgrjasBpARWYu/Oo7MIAwAEIxGspAAYMEEAEESgwgL5DiOz0AAM4YDUGGLzMQAIw Z3wCAtoGkHIKvSLQL8AACEwsxDWYzfHIFI9dA6wGG/Av1VVfnbXYssotA8HvdmDB1FWX/MDhh7/7 gd8qq8AyCdWKILMJky9OcOSWQ36AAeKOWy2n554r6v/jH2jbAgOWJmAA0a4ecADSF1wwwAURSOD0 AkIsMACsUzuwd9jOJqA0zIubkLCzBojNwtpnO+1A3r5jAMDLbMPg9t3PWx29tglUbwOxrtaNNwbJ h20w164zHgMCiD9gQQAOBGBB+/Qv3PgMk2Ne+Ql0m2t5rAaLVc4GmLPPKSBnohMdqlK1wHYZb3UJ qBcUUue6A1xgArK7G+5+oDti4S1+YeOe6+Y1vBSAbXUGcID6UIYAjlENhCEMoQESsEISQCxtL4xh DGdYQxUgTVvhG1r5DqAtBpyvgq/LAfvo9z750a99FpBVwiJ2sPuVYH8kCJsJiMUrAGQAbgDYHQB0 9rn/MiqgUp47VwL8N6oGiMqNpiJb62QVgNZZCokHiADVALZBHizgX2GkmtaIqC0ijtBpdSReCXIF QRWyoIURgBUMLWCBXOlQWz1soQIkeclO9jAFCgOi0oi1OSIOzFl4/KQJ6aeBw1HyifTTHd5miTdV WrF0QFwdygJQsQRoC2sVex4Ox2hAAZwxADobl+TWyMY3pktVPkwAA1wlxFQGC2II0KAf/eXB4VlK AUSc4eq+OQB5+W0BB0DWI8WYgA5MLwAL4ICldqfD5K1Qk05LWCfrqbwZKMxV2hoe3Q4wzTC+zHUG WMDqbLnI+SGula185QM4JYAnRsyIiAzA7m4JA9Xp/3J6S0xc2EL6AAQI818OIGMABLBSimrAf6Da lBttZroVkJKgr7rj5mB1OAD09HC602YOgrrJqh2UnJsr2iHLOUMaGq+f6xzaPJvIAQhSzQIAzSVU jTeAz8UPAb7cZ9gAylCNiZGKsZpVBiZgANSVUqFOlcECJtCB9lHUfQ6F6BMtcLTFHUCQq9sqR1vw OJKK1LAlhR/HUnrGM5IxAS8dAbsq8KkCVDZoKkAn+vD4KgBA4KcP8OkDZCkwHTDgbkZV3VLhWbHW Fq2PIgDbJ1sIAL5CjAO7e12sWqjVOprzawo4nAAgwFKx7vC3MDjtApjGNA/kjGloDVZCZ8AACRi2 ov94PVzO9uos3CFNaQjF5GBdgDnDeiBx7TuvSJ93N50ZoFqQVSYJ3HguN1qWXanaIupQWcG2ivaz D/DAZ2W1AMX6awCwpUFQcwjWcZKQlwFgwANaiTgEbC4BClPY1gRrwk3yNLRRfN3hJACBgW7Ot66y nw0nOlEBREAA5TMuig2gYhboDp6s1BlQmevccTEXnTyMwVwlMAHEFs2hd61f8JYl1QuPcrwtwNwC 2tfK84a0yolLHnsPOFEyyldyb4TpqeJ4AoXeMV7UBLAEPWCvED7PX8UqAd1ccNoXGtG1FSti+5QK UEHWcbbc1K4GFACx7mn3vLBCaPJQGNcR3FBXHsj/lIzHemFVnhbHsHxABab8UB1rGsjTTW4EtpbO PcfLobC0ANdUt0FYte7J441WCnTGLU4LV8chTTJYf3XaA+oMspCNWRuh2QIAeNRZddSWgOfH5ma3 OWxY2+MA7AerAnPTxgd+2tD0TOHEmo9iBUZhEHflgmzu7tY9PQCSKdzkRccrZdn06bhCC4DiThrN gFYAPCsAywpsmt9PrIBUkeuuCQRwbe/i66kzXdLNNjpWRCzrojB3gspJmMpAxXjijiarXmvgAMn8 FuUsS1+bti6ryn4ABJrt7He6eY8L27ZUOWyCP1KtoIGNcIXFevJXyUCTE8DA/DgVWvSxj+hFD2+E /5PXaE2yz9OwgvGkffuyTE4tnf7OetaHtgCtb923m/N5ubc2gQkYG4lZNYAFDPuuEYZ9bNOTOKL0 578rmouGt3riAjD8RLBy7Z0E9TWnOkVxYedXjuIOoYCdXS9YXRLmKKu0bm08tbABuYj67CQDwE7N nyvAARMgWgLSjT4aIu7srtt8ABqw6IZl0wFDxy69XW5cgIp9BeYGL66AlWjX7R5YhRS3KhFQgbK/ ir9K923F+FpKRcP99szKH7awuMUZ2nq0wbv+3n1Jw7CxVwPxvZYIXEfxVDmw+okPKMvtFSy+6Yq9 MQ+s4ze5ApvDEJ7pfFeBO5lOzkO/3J83TrBiYf9I9Hewsmqu0wBGpICdF1ufRzT080NTF1j/hzDO 03wElXrh5VYoh1Bv9wLZZHxnJ25olkuc1V8VODRyNyjUJwJaVAKvszVcs3vIR4PB031Y4wBO4znf Ei7khy0iIAAtyCs9FzYDwHhUtAEbIAFM2AAAoIMxJ0Skxkv193n352rO4gBFg2bysn89V1av1zoI OE4jtGqbtTkKqIB/5mgPaHrpFnYxtk+us4aPZIUBYClGpIExRlB5mHrTJHwg+HlsxS8keFxc6G4f SAJxB2UzU1Mk8DpmaIZ4dEc3SFD7Bz86iEDgElPQMoQoY0cxdoT2wmYRswENoIQboAAS8ITT9on/ FmZs5AZKeRNC01NKWjiJ/EJpiWg9RmVHJziJCGU1DMCAt9dCoWeAxoY8cSiH4ZRJVRN2SIQ6cdhW fsiBzgeCs5grcMhz8VJH16iIFciIWTR+Q8M1+6VTeISAR3OJ8HNAILcDSIOC6tdyuvIp9qiEqwiF kddZ03NtoGSHYQMADUZEvEdgYQd2uxiIDrBoaIdmwAiHT9MAdPgBr5c6kriNPDeH4Wg8VoiHRiSN 9eRWGjlODegudrgAJUZIE0iS0FeLK9g4r1OGDWk0FbRkEMCOOjhGIuc4nJNA/OOL5YNg9Ggv9liU n/KEhMYrR1NBG/WPVhNDLhdISmOCQISCZZU2/zB0iGPlkBwnRK3HKwFYhpuFcp1kRxO5Klboiyd2 SW1lRL5EhiVZh0+5XCXmUYdIk+FmlbxyluLoQ/KCYejDhW4nTZ4VATi5OzvJk2CGAkgDl8lTTo1X j/YYLFZDY10Eia/TlCigUE8ZkLo1PSU4hyb4jUKTQjskY/JycKDZMMTCejUZPAZzf2w5TnyJMCkU P1zIjVx5APo2fLcJT0xzk7OCfHhkMNOkl7BSTknUl8LmUTbzQwR1U8CIYbMCAUyDk5vUk1GWmFuE gQcplKQIK/bIZur4PBQDjjRkYTQ3AhsDQ0pFezqUNWq5UH/TL7IpY+SjW7S5nItTlGZoMAvQmf8x tDe+eIeZZJ8ydJcoJ59kOC/r+TUIulxMUwEQcJPRpVDH6YGxIlDMeQLOuZdtlVAXWqEV8GPe50go UHhBiC2ypoiu5nYHSUPh6QFHeZEHYDUJ5lfIVkMB6gA05JKo6WBzmKPlJqBBejRKNYcqhlazMgFG Wk9kSEQ1lllPSZMPmXZCGqDwBAM9CpwRAHD+9qUUWqElRjWMNqTGBl4b2aFyZjRU41lkGqb+djgl apgh5DU3kwKF53jcZGFjaU/hCSwZcIaEhKdKaU/FpnrgBZ8DipCByZ8xQEpkWZZEk2hwCakf0AAS sAHDqD1PCm2jqaEMs3lk+Hu7R2rQSEEW5gD/6TSq2rJc/AZgASeVEIQ8kDg8a8qcKlqLSfOmAPdE n1Wi7Ngw1OeINgVemxNA6sh04emEq6ZsHgBERIpOVVR/SWOgR0pBBGkDReNqCiqGFmZHqESFJWCU U+eoSio0ZrahtKQ0q3MsMmkpAYqp1vqqEXA4ESCrD3CvgbSQ2gpPqnOH9MmmJlB4BSBEqlNL+9o+ 1omvl+hf4VKs27I8w2OASUMsY+l+sfJ3bOY+HuBfX8NQa+N4uhmvA0sD4FNbqpMAUXRNqOOvFHSy 8/UpK+lgtqpKKkhDeGN5CcOOlod6zqJUjbY80wScDFuh/Fo1yjmWMemcL5koLSouQCg6xiZO/wqb rxXaPnaaSMsSLi5AOoqYOnyKNJukjkqTAdVmbOqFaEFWcxJXtcuYoA0amEOLsgngABwAKxbQrp5q s5VGNgG7T7tJtzh7hzTUjQnTQlqYuAUqNr9Vi3UbTa6yXLAkSwMglpU2cPTKUSoqbIz5lsJDNdeH OIYJUHHliSYAtkRYQXf4mYZrtvz4YVjoNwJpt7aXS1YqiVWnA981S62raPGyaky3QnCLcrtpkcPL MN74oxhmUqK7f8+TYc+KNHFVTZELuJMrpsJqn5eru9Ubi7E2fXXnuZ8rL3hzK9pbugTngitFLVNb bI7JZM55kUoDMaS4vjYUZzNQvFYqlsp4vf9zo6Ao2FQnBsByZpfHa5EFPFu7RUM0hCzVVkhE9EGM pFFbJZDSZMBb5FEH1y94I4n4C2XSt5h/wzV1NEsd/Jdjoy2asqLZMrF+iYfklozLa4Y7q8KMmavF xjVqGYkFrMG2q5YWiTxcyzB2mY5jWMTFVnnUGSzW1rqB9TQB+XcHbFQPusG+NEusCsIvA8RWNIQv KDRHTEtcW2/vW7DTZ6xyNMH+KCsIrKAYRru7ewNu/K3BmytAgMFaCUSJ5D0uoMe5yccz5Mc+9FeP R8acBT+EGV1TiW8Ms3d23IVPez9hPLNByJ0by4U/ejlnXAKPQ3J/PEMIJTEHCIdGE8cmZKB78Lh3 piyG1TosDXaX3bOCgHyIsyzGD3mlp0xIuTyJVKyuyuiQP0qwnhJTndtFyRIDpOO1LsyYF7NF0eVD hjvJchbNRxBd1IzNM/Bo2IzNEdfNujKlf2zNxFzO5nzO6JzO6rzO7NzO7vzO8BzP8jzP9FzP9nzP +JzPdhACACH5BAUeAB8ALAIAAgAHAkIAAAX/oCeOZGmeaKqubOu+cCzPdG3feK7vfO//t49wSCwa j8ikcslsOp/QqHRKrVqv2Kx2y+16v+CweEwum8/otHrNbrvf8Lh8Tq/b7/i8fs/v+/+AgYKDhIWG h4iJiouMjY6PkJGSk5SVlpeYmZqbnJ2en6B2IqGkW6Mfp1kEBEKrdq6lgyZWqaljJUOzRCO5J6ge u8C/oyi9vMHHsUentkXNRKuwH7DSc9Ssyn3PVNu3wkLCtd/DRrbm383p6Ou72Urdv0jS2NV69e53 z8Th4yTg/uICrks2bNw/JOJ6xQvGUOHBhg8dwvtH0N9BZgSRDWS3sOCyjAwxapS4MSI0bEWu/w2J pnIaSpbyop2U6bIVvZv49ujbB0zkwoQde/J76PMjQo7x1EEMavCcM34Gn5okatHh0p9DsSYp+lSo 0HJZuXY8eaSlWZw108a0yTbt2Zx6dPmcm1VrQ7pTrxqr+HWp06l/reICG3GwOpBK7Yq9CzVsXcOO pRq55xbt28plYWLe3BJuvqh4FXNMHJocYY9Jm/a1OhYo68TITof+uy2Z3Me1B8uOLNoY65UoyXK2 fFOz8MvIg3ums5Oqc6aAecN+zbsgX79ITU7/3dvuc6Njv+ed+Hx23bwplbdNvl79ceLt2y7P07x7 0fv2VYP3fr+ff4jbufafZOYxttVj4oUX1P9R5Umnn4LywTdcfK1MJiF7as03x23fVUWOQHuRRFFU e3G1D3Vd2XZdiqqBZJ1G572Y4kgljngUPKVBRpVrFs6Dlks0AekjZUK+ZBxLEmqo5CfkLWkFkVpA 6eSUpDS5JJJYZqnlllx26aWWVIYp5phklmnmmWimqeaabLbp5ptwxinnnHTWaeedeOap55589unn n4AGKuighBZq6KGIJqrooow26kgCBzQhQAUcJLCEAAIwUcCmnFLgqaebfkqBEggQYYABAKQKwAKs MtAGA6wuoKqqUzIAwBCwsjprqmAA4KoQucq6K7C3OvqGAKlimqkQvEqhwbOjFvBBqA1QUG3/AZ42 gcC2p56agAIiAOBBBA7AmsYCDAygwLrr2uqrrL3iui0CuxYrBQPzbjvrGKnaGoC67GLArgKwpgpv FvaiC/DAFnQwL7PGttEvpka4+8SzGgigQQEVSCsEp6FCAcABCHR7QAIAGDBAABFEoMAA5pKh8LoD JMBAAgk4YICtBiRw8Ba8IpAuuwC06yu9TpQq9MIDE4z0F76e6q8DVOvsAAZY24rzAQfYW0Wx23Zg wcsDDOBAww+knfa2H3gd8RabJuHrpVorgbGyGkybAKdUMHByAgaQfCrXKV9wwQAXRCDBywt00awQ CywcAKQJrAz4qSn37LYVRw/9sgNmU401/wCTP61EqgCbHfrVGARQ+uZXpOy61CMb4PoBrHcN6QGo wj4FAmo/YEEADgRgQfDIl/q2F3GfOq3Ht+bLbKr60psqxtjfjakCGD+/d6eihvrpESWfHK4If3N9 wAUTHE5z41rMPUTkNNfM9beW88774L5LAQACqTOb6wZIwABojlQDQJ0AC0jAAwLNAAiY3cqiFgDc YUB3XPNW/6AAPOQNz3jIC54F3LYABAirXssrAsWOIADnxS16waNXBgLQwbW97HrKyqEAFCCAAzzr Y9P6wAE41YBNFXF8SMhc1xZgwMCdrF7WWxf8OFhC5UFsCDY7ANkOUMGTvcyAOAPc/rqWBf8ARiBV xXOdBSwgKwa6zncAXJcJ3UjHDUqhZLb71+T6hbuu5U9qZLRCDdOmgbStMYQeTFjZFsnIstkRUXHD VABYmLciDHJtNAwhAmgGgIxtb4cCCEDGfhjEuD2vAA2o1qhOZzPM6Y9wAEhbLB8wy6IxjoMOUB+v 5PcByimAixB0Iv68tbVuBe6RSwBgqhLQAdIFYAEcOFkCGWi7zQFQXWZDQALoWMBqauGE9VIVqzIw gZ2N0W/IRMICjqe2QhbykA/QmABCaAGUTW9WDGDAF/+VTkKZMoMGMMIKhXBJD6wteAZdG8Ac4MkA hBJToxxCAyTqsYGezm8HYACqTgZLCMj/8qMPoN8tk5bLwAXgVvbaHf54Vz4vDoCYkDKmA3+XwPsN 4IMcCFzlbnqqAfbUawCcpwDSyM1uevMKkVtAy1rmAYy1LJz1sqcUFjCBDgRPnsJjpzvp+UQjjKxs BuRiPw0FqQ84T4XLglzwCmnQGrJ1baBbF0N5yENPJqCSzxtVKlFZRGwtYQHpUx/JZunRB3jAo6ta AOgA9ittVbBnpvtAS/FnwHxZVpvqm92pphgFZVpgsByo6awAqFnNOjCOGqhABIrqxm5JtQoMeCYi 1aZUpjq1ZQvY7BQYIIFLzjOrhHwWPSHFWWaprItHTeEQNCaEARrBZ1fN2ANqiNXp2qxV/wPImAEw dVdSPq+I0+rrEUXlVb/tTn2oKuz5PCCuAS52pE1w5kmJwK37fXEBNmvn2rLIxdnx7rUcVICqZDnC rqVNAhDgI++aeCrlbZKH8YQAa43aYCtETrazrcACZvsADf93rB+gqgQmcMnpGoCd1Y0h5TY3sgxW DsSHWqFzK4a8QoZ0rWpDwBNXxYBCylNjeB0CKj2mSmxl6wi55agxD3s89jq5vQR8b2PbxkuvpipS Q7hsvrjGgODhd2sN3F9xJfs4I8RxljkGQAIueeX9hVVzqK1APFs44TDP9F4KeKacQ1gBDe8ZeRVo sW6hwIAI3OwAG85xt9iJyHrizABjbv/bGBOoXCjgbFUhxG+iVXxp0GX0WQK4612HwLXvrjK+l4OU AV1nUAg4+dXyDQDWyvYy6zFRXRusnfIi12VChtSJGcWXCdG1YJ/uD8AIoNp8yXfDtWpgZIx+wLNa vGBj1usBEaBznU2LKirEcQCI7rO4xT2yBYyb3G++czIn4K6jbeuzi+awNlUKu5GtGsbKVfOmQ0q5 fX95j8QDmAZ8CGRSY1m8RUxrEgfX0wEe9snhShUD32vvFi9bbt4s2gV6HdIxCvZ+q8qjZvdnLxPq OLlZVtcEMMBojW2NuuzdXR57irL/qUpjDd82zfG9ybLlMla6ajOigd7Gx5L8CUJLwAT/JqBmwTb8 xJfc1v2OfgTSYW55Z21CpHx13Vidd+is2tqlXRdXUHtXCBTbq5GzNamqM1zkA4D1kyVer7Kh7MNd e2TF9Wm4XAHUZJRDmc0aGLirE/TNDrCmAhwwAZIlIG0C8IDYZ+ly9BrTgACYQAUmkPgri7zOPe32 HbcIbIxi1HanF3kGqa6tzZez6U73KQTn9dlXsr4IKQtkxFoYUCdsHWc3C+PHORr4jDKR7OrSQHeF XACLrv1TCh+CEnMed3GJi7272sAGJMD9BqQMZSaVuICZYG+YtS+msge8zzJIeNbLSqeJN/PinUi9 j196mVsrtkkBuPQOdGDpV7Zt/oV5/7/zORVkM6ZXbBWUTxgVbG/XT5u0dKgSU/tzeU30ccBmeLh3 TCnEXEiQdULQNRkQRvlXgcKUfyjDRFjjAC9jdsz3NW83QNUXcdnXANq3AQogASNTc7xzaDQUX++i dLZzecZ0MkmWcxeogSGGaJGVZfOnU8UkWCRYgnlEMvw3AVPIdG5WZ1xDgKNXUsCWUQpoQKaXTwB1 e8m0eOWELoWHhAwmU0koekSQe/gmKBZlKr1Hai1GhUNoMlIIABBwfMTDgt0DBabkVYOjgHEHa7uS SqmkfTr4RBCkO7TiBAAwAPilP21oMiHXLd0EUG5TceQjQImIgcNnea7TUlBIgn/Hhf/AxHOhczIM 6DefNzsN2F9OpITJFDrPVDtuSGFDCIobqHsdqHCHOASTZFxO9GUmCGyQYisQEAGCSDzc40NRcIxz mIF5hIlzNyuO+I3e90S8oyr/Mn7xtXiwcopDt4VGJYy4J1bypzNhZXl+OHwzp2PFBFMyV4t05ERe 2FmLpzNhyI+2yIAXiIakEpDPlGD9pXO5KIdts0cpNGNFgI0A1Ta2k1G+oo44EyvRKI1RJmBnd4ci U4rbaADdKC7fuCt/Qy/iyDWU1gQLoACDE3SJ5Tcrw03aaHI8eTMXJ32Ll36eyG1DGVYM0ABmWIFO 94vU5I8QiXQKCYc66YnaCIEKqVT/CXY5MlWEiDaPrGd1dQgoFKkEZeU8srMzkAZVqwIBEKBauBVl CZQxFbBc0YeNGKeN/8KN9eKIADCCwocz1oNBl5hOuQWGMshIRWWSp1I2i3k79WYARDU7AhhWjpgu JUiPE5aI/8hBkBmZk1mKeQaBnflMLROIYaeOGqlRlodSl8iBlfZcWHY6fagyCcSWbNlnEdBnaeOW 08iCtzJQd2iXSWR7blYzNAgAjih5wnc/+IKRKKNjP6kEClM8XWiBbtQ6pag/1mNC+1J16eKZE9Y6 B7kr+WidDCSepVhBPPedhLeVOYedzhiaSMeetZWbbAlOu5Jbqrl6qmI5YfmaEZlB/4sEAH+GPB6l WoJINW6TSljARKoSU3lkfaniAamkZlRohNLHUdH5V8o2mQe5O890Rx3qoYOTAA0gARtwlMHnoQfZ hVY0VcpWj6f4dMKEaMUTaUmgWDfaMnuGm31mmwnGmP/VhbJSOY/1n6+pRLwzoNgWPNGYNiBJQLZS BAyKh1Egcbh2craTANiXWH2Zf/41pcyyYCDmK5MJn/dDgTBmpq4IOB8Kjv9WZ+g5kMR4pbEFbEQH dIfmcS/XRzjqna6jVHJWWHx2iaADhfqjZpbzlABKBMJJBMjSRUvqSNhmm8ETpQMEAI/6AZoFqdGH eyuTQe7ipielKt4XRqzmAWHFWf+AVWZPkFufeYJE6m0ESU1rRoF/Y6IVijMkWqNc9KL+k2T92Ugr EzjBkqb3o1h1+ld5pFRQSqhNeolCCqb4lZGu2ahG0DwBdYyTUjv5t0j7pjaYSoaa6jFH4DoC9anS Zzn3pzJRgwFRhTLsJTwesDNAFZYUlJmJmH/Xaqfu6Yea0y+reGnMolEOKav9eqWUuEgDVEKs4kZF OnXlA2BLQEHOqja2GQG0pDqImqgnwz/YSpJo50Lm+gEUo2Y9szWLBCu5iZvTqG5K4IHk9zdYmjIC dgArkwEG05cJZVDf5zU/QwUoW6sMpj7CB7Mi82gJMEKjpS/dUjLFBGAom5j1GLX/+HqAmWM7JpdN JpeeJ6VBAUqxrLRZGptpi5SmMcWDiQqgkaSuJhtkRWCW25SyO1U27XanOwcFIut2mQVLXAeTvTNg tPSzuAesUzC0q/aGzih8Ehk7ALBGxFo1JuMtW8piWilyXIm2lfs1YVVzPpNsZfOwAeewqEq4kqZT /TS0guqy3wlurNi43fmam0qXH2gAu6qPOIuYxiRVcYOub+uoJfsBVbpwq8kAjvatNCShYEtfU8Y5 l5uduQqhYiu0lZOdRXi0jetVzwu92Du92qIq2qRmqzIrx9eFi+QA76dHXvM/NuO9VXc5d8tIrLi8 2CoFzleyDGqhXsmvx/ZavZuM/9nqu0LAoDKbjbKoKsMTeGHEsMQUioz6NY8WhlO4YAQLNFppf9iL tNJ3wVKYwe5bsWQzOTZpMHq0ev8SW29UwRu8QKnLq4yUS/M7OR9cv2hXSWP5gkm0vTJVc0kwvERw O+lKvHkpYBH8rz0TtBE5w1darUbsLUiMMNq0lT4FRk1YdVEslasGWXX4VW5ErJLKfg5gM2p5idyW ukzcxD6DpLu3LDdsicu0w5XYwxNlBINzpT2TQeILodf7xBG0rAiTiVu5Y2LAvv+6ZmNFyO5pyM6L mjMKeP3FyBhcpoAsU4JMw0swu0JQVkyAOShkBW3nP7E7xlWHtWEgymVgyqAMVRFlNMasHICtPFpC q5aWjAQhAAA7 --------------Boundary-00=_RU81QL80000000000000 Content-Type: Image/jpeg; name="BackGrnd.jpg" Content-ID: <33AA58F6-CAD1-46A9-B334-261C354B10DC> Content-Transfer-Encoding: base64 /9j/4AAQSkZJRgABAgAAZABkAAD/7AARRHVja3kAAQAEAAAAHgAA/+4AIUFkb2JlAGTAAAAAAQMA EAMCAwYAAAHbAAAC1gAABZX/2wCEABALCwsMCxAMDBAXDw0PFxsUEBAUGx8XFxcXFx8eFxoaGhoX Hh4jJSclIx4vLzMzLy9AQEBAQEBAQEBAQEBAQEABEQ8PERMRFRISFRQRFBEUGhQWFhQaJhoaHBoa JjAjHh4eHiMwKy4nJycuKzU1MDA1NUBAP0BAQEBAQEBAQEBAQP/CABEIAGUAcwMBIgACEQEDEQH/ xACAAAEBAQEAAAAAAAAAAAAAAAAAAQIGAQEBAAAAAAAAAAAAAAAAAAAAARABAAICAwEAAgMAAAAA AAAAAQARIQIxQRIiQDIQMFARAAICAgIBBAIDAQEAAAAAAAERACExQVFhcYGRobECEsHhMtHxEgEA AAAAAAAAAAAAAAAAAABQ/9oADAMBAAIRAxEAAADtRZYE1ASghQFgUZoCkKSwLmhcllAEqkSkqFAl hUomoAS3IoJqFlDNpFEAQFE1AIVYAWIVKAJRNZpYCwVmmshKACA0CBAUCBYGwf/aAAgBAgABBQD8 B/yP/9oACAEDAAEFAPz6/or8H//aAAgBAQABBQC2+ZeHjbD+saX6hwXeDW1Rg4xLLTa+m7ZiIEsI 1MTiHP1dYpvFADiFM1/X6nq9byuwdPPz5oFofWlEMQ9ULKrWq2ppG9Y2J6INQma9lVTRdlUKgHzX XSEECw1SYu5WsGoJPkisZYpx31GvXZQ/JM3VwShzVTsp1EZbBI8LcaUSih86+s2Zl4Wp6+lAZnVs Dkjdku5m+lJTdXDG2SHM9M2wKX1YxsaZTTwmoVrYnqsMrM652yjs01K0mtbGAz6Y5dpfqNz06qpq 5QNjiIjiZtbhtceNuf0jyeqGgu6rXMvI4omPWbPMYzEfMI+axHnFvOP4/9oACAECAgY/AGP/2gAI AQMCBj8AY//aAAgBAQEGPwB72Yucb1BfIhFEaeZ+xRXFQELN+HEUQdjU0Xn4g9gRCQcpw1yajGYs P/kFvUzvjUBWrIMFHI2OJQNEAjiEEFdTmfG/MTHq5RFOnpTV3kzCBx7x4YOD1AV5uYJvnqMA0hep jfwpYCwC4Bx3q55zeZRBCw9TkoIuHw78RdczSNH2mgqcLpRC+RASAkA3B13mcYd5mR84c/yOx4lW tRAZ6mGDhiP9WgXVyhWA+xDgMOWGMsTg/wBTz8SjjXrP8hHIlX1MZ6mDzgc/cIV/iyN1GBR0MQMK jnEzvvMz8mUkErKlfqU63iV+IKNH7mNZBLFQEpEDeDOV32IVn8WR4caoywqI2p695mbZzNUQIcKf k0bo+0NpCqn7CiQiNGXkdQen1DpjGeZ7WNw3pK+I93maCPc16+Zkf6XxMCsFwAkaiIB57vc/IAhZ /HqZBBbB0ZokAEOGxsYqBgPp8agQBu4VSMJdqx6SwDsGBrTmAR93uZGX6KePowEADAIjoX8gw459 CICaW/MLGvodQfkDW71zBxRHtB3j3jC4PMIYoAgKNfPMCQNN7jCzvlzXPopzhQvNZY3CRya9ZrEF fRE0iCB5mscZuVYfKmAi94uE3Q8qfytQ7xD0svmFcmaxNPI8iMjh3pmF2HbzqeUi+YkiD/MrOl5L mbwPuWVfmXpv3hDH8qAjPpiZHXkRnSd6ZhB53mejzKV6US0K9TCCLyCeIhtETX5MsHBGJkD/ANiF kMCE2qGoCdZ8Q8AMGpYFqEhdhRIYH3CF3d1M/Mexma+4CwdQ2Ddcx0exAlmj04QUQd8QWLB/iB5G xmEg5TENVZqPYzFV8eHAy9T/AEc8a4n3Ov6g/VwvE6lpQ4VNysXzhS8esOO8w/rlF/rypjV3B5H1 Knr8T//Z --------------Boundary-00=_RU81QL80000000000000--