Question marks in emails instead of diactric symbols (polish) in Scarab 0.21

I have a problem with e-mails that Scarab (0.21) sends.
They look exactly like here (some fields are replaced with 'XXXXX'):

<BEGINNING OF EMAIL>
 From - Thu Jan 10 15:16:14 2008
X-Account-Key: account1
X-UIDL: UID3904-1130357794
X-Mozilla-Status: 0001
X-Mozilla-Status2: 00000000
X-Mozilla-Keys: 

Return-Path: <XXXXXXX@XXXXXXXXXXXXXXX>
Received: from XXXXXXXXXXXXXX 
<18171307.1199974562355.JavaMail.apache@XXXXXXXXXXXXXXXXXXXXXXX>
Date: Thu, 10 Jan 2008 15:16:02 +0100 (CET)
From: Scarab <mail_XXXXXX@XXXXXXXXXXXXXXXXXXXXXXXXXXX>
Reply-To: Scarab <mail_XXXXXX@XXXXXXXXXXXXXXXXXXXXXXX>
To: Maciej Gawrysiak <XXXXX@XXXXXXXXXXXXXXXXXXXXX>
Subject: =?UTF-8?Q?[G=C3=93RNY]_Zagadnienie_#GR1309?=
Mime-Version: 1.0
Content-Type: text/plain; charset=ANSI_X3.4-1968
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by 
host17.nazwa.net.pl id m0AEG11Z005309

Raport aktywno?ci w?aczony

   *Bug GR1309 - Test*

   Odno?nik do Skarabeusza: http://server:8081/scarab021/issues/id/GR1309
   Modu?: Górny


   Aktywno?? wygenerowana przez  Admin (admin@XXXXXXXXXXXXXXXX)  przy 
2008-01-10 15:10



   *Zmiany personelu*
   - U?ytkownik "admin" zosta? usuni?ty jako "PrzypisaneDo"
   - U?ytkownik "tom" zosta? przypisany jako "PrzypisaneDo"

-------------------------------------------------------------------------------
<END OF EMAIL>

The problem as you see is that every polish diactric symbol (like 
ąęłżźćńśĄĘŁŻŹĆŃŚ except ó) is replaced with question mark '?'.

No matter if I set scarab.email.encoding to UTF-8 or ISO-8859-2 - in the 
whole e-mail only "Subject" field is properly encoded in UTF-8.

Is this a bug or something wrong with configuration?

Thank you for all replies

Best regards
Maciej Gawrysiak