Instances of two subclasses, Error and Exception, are conventionally used to indicate that exceptional situations have occurred. Typically, these instances are freshly created in the context of the exceptional situation so as to include relevant information (such as stack trace data).

A throwable contains a snapshot of the execution stack of its thread at the time it was created. It can also contain a message string that gives more information about the error. Finally, it can contain a cause: another throwable that caused this throwable to get thrown. The cause facility is new in release 1.4. It is also known as the chained exception facility, as the cause can, itself, have a cause, and so on, leading to a "chain" of exceptions, each caused by another.

One reason that a throwable may have a cause is that the class that throws it is built atop a lower layered abstraction, and an operation on the upper layer fails due to a failure in the lower layer. It would be bad design to let the throwable thrown by the lower layer propagate outward, as it is generally unrelated to the abstraction provided by the upper layer. Further, doing so would tie the API of the upper layer to the details of its implementation, assuming the lower layer's exception was a checked exception. Throwing a "wrapped exception" (i.e., an exception containing a cause) allows the upper layer to communicate the details of the failure to its caller without incurring either of these shortcomings. It preserves the flexibility to change the implementation of the upper layer without changing its API (in particular, the set of exceptions thrown by its methods).

A second reason that a throwable may have a cause is that the method that throws it must conform to a general-purpose interface that does not permit the method to throw the cause directly. For example, suppose a persistent collection conforms to the Collection interface, and that its persistence is implemented atop java.io. Suppose the internals of the add method can throw an IOException. The implementation can communicate the details of the IOException to its caller while conforming to the Collection interface by wrapping the IOException in an appropriate unchecked exception. (The specification for the persistent collection should indicate that it is capable of throwing such exceptions.)

A cause can be associated with a throwable in two ways: via a constructor that takes the cause as an argument, or via the Throwable.initCause(Throwable) method. New throwable classes that wish to allow causes to be associated with them should provide constructors that take a cause and delegate (perhaps indirectly) to one of the Throwable constructors that takes a cause. For example:

     try {
         lowLevelOp();
     } catch (LowLevelException le) {
         throw new HighLevelException(le);  // Chaining-aware constructor
     }
 

     try {
         lowLevelOp();
     } catch (LowLevelException le) {
         throw (HighLevelException)
                 new HighLevelException().initCause(le);  // Legacy constructor
     }
 

Prior to release 1.4, there were many throwables that had their own non-standard exception chaining mechanisms ( ExceptionInInitializerError, ClassNotFoundException, UndeclaredThrowableException, InvocationTargetException, WriteAbortedException, PrivilegedActionException, PrinterIOException, RemoteException and NamingException). All of these throwables have been retrofitted to use the standard exception chaining mechanism, while continuing to implement their "legacy" chaining mechanisms for compatibility.

Further, as of release 1.4, many general purpose Throwable classes (for example Exception, RuntimeException, Error) have been retrofitted with constructors that take a cause. This was not strictly necessary, due to the existence of the initCause method, but it is more convenient and expressive to delegate to a constructor that takes a cause.

By convention, class Throwable and its subclasses have two constructors, one that takes no arguments and one that takes a String argument that can be used to produce a detail message. Further, those subclasses that might likely have a cause associated with them should have two more constructors, one that takes a Throwable (the cause), and one that takes a String (the detail message) and a Throwable (the cause).

Also introduced in release 1.4 is the Throwable.getStackTrace() method, which allows programmatic access to the stack trace information that was previously available only in text form, via the various forms of the Throwable.printStackTrace() method. This information has been added to the serialized representation of this class so getStackTrace and printStackTrace will operate properly on a throwable that was obtained by deserialization.