The proper sequence of action in a 3D Criss Cross Assessment is as follows:

However the test item of a list is determined, the essence of the first step is to list a list. This can be the list to determine a test item or an opposition list.

There are several lines in 3D Criss Cross. Each line is derived from a test item and is thereafter continued by opposition items. Lines are lettered. Each line is an independent zig-zag of opposition items. A line can begin by using any terminal established in old Routine 3, 3A or original 3D. Or it can begin by a test item derived from an arbitrary list such as Dislike, Like, Who by O/W, Dynamic Assessment, a Pre-hav level assessment on the pc and Who or what would________, a list of withholds or outflows.

The essence of all this is that one takes a button and pushes it to get a list.

The List is always derived from the pc, without suggestion by the auditor. It is the pc’s list and what happens to it is up to the pc.

The auditor pushes the button and thereafter is an interested writer of a list (while keeping the pc in session).

We do not care how short or how long this list is. The average list is about 25 items. If less than 12, we consider the pc is ARC broke. If more we only know that the “can’t reach phenomenon” has set in. In the “can’t reach phenomenon” the pc keeps listing because he “can’t quite say exactly what it is”. This is an actual sensation. The answer is to go on listing until the pc has expressed it to his satisfaction. The phenomenon is: the pc couldn’t reach the right wording as it is too heavily charged and only by giving more and more items is the charge bled off and then the pc, able to reach it, can say it.

The essence is to get a list as thorough as possible without putting the pc under a strain. Pc must remain interested. Forcing pc to list more and more and more when he’s had enough wrecks the value of 3D Criss Cross.

The list should be numbered, should be on legal (foolscap) in two columns. Readable. You don’t recopy lists.

Date the list, put the pc’s name on it, and the full question the pc is being asked to get it at the top of the page. The back side of the paper can be used.

Additional sheets can be used. But if so, name, date and item from which list is coming must be written at the top of second sheets.

Numbering the items has little value but it may be done.

Do not keep pc on meter while listing.

You will see a pc getting dopey or drowsy while listing or nulling. It is good auditing to run the pc’s havingness process each time you notice this. Nulling is

accurate even when the pc is anaten, but things blow much faster if havingness is run.

After listing (or during listing if, as rarely happens, pc goes drowsy) run some havingness.

Put pc on meter while running havingness. Test havingness process each time used.

Assessment in 3D Criss Cross is aimed at straightening up the bank as much as obtaining items.

Lists which won’t nul on repetitive assessment by elimination have not been differentiated, or the ruds are out, or the list is incomplete in that the wanted item isn’t on it. A 3D item is heavily charged and when mentioned discharges much of the list.

The essence of this Differentiation Step is to read each item to the pc and have pc briefly explain how the item__________(whatever the list came from).

This is done easily and in a friendly and interested fashion. It’s the pc’s list. The answer that must be ascertained by the auditor is whether the pc wants the item left on or taken off the list. This makes the pc look. And it blows charge rapidly.

This step is done with the pc off the meter. The atmosphere is easy and pleasant.

When the differentiation is in progress pc may want to add to the list. Let the pc add what he or she likes. Put whatever is added always at the bottom of the list.

Pc is taken off the meter for this step.

Put the pc on the meter. Make sure there are no session invalidations or withholds (as different from life invalidations and withholds) and begin nulling out the list.

This action is done in a brisk, business-like, staccato fashion. Each item on the list is said exactly three times with only enough pause to see if there is an instant read (about l/2 second between speaking the item each time). The auditor then acknowledges and says, “It’s in” or “It’s out.” Patter would be, “Tiger, Tiger, Tiger. Thank you. It’s in.” Mark.

“Cat, Cat, Cat. Thank you. It’s out.” Mark. No interval between items read except the split second necessary to mark.

Pc is expected to be silent during nulling. One does not consult the pc unless the ruds go out. One answers the pc if the pc originates but then only TR 4. One doesn’t enter into discussions with the pc. If ruds go out all will go nul. If this happens, quickly pull session invalidations or withholds, and get going with nulling.

If the item clearly reads in any one of the three reads leave it in. If in doubt leave it in.

Nul with sensitivity at 16.

If consecutive items which have heretofore been live vanish, suspect session invalidations and withholds, clear them, and pick up the earliest consecutive X where this might have happened and carry forward with nulling as before.

Treat the list as a wheel. When you arrive at the bottom begin at once at the top.

Use a slash mark / before the item if it is in. Use a cross if the item goes out. If whole list goes bad and you have to re-nul it, use other side of item (to right of item), then use a different coloured ball-point. Black for original and second nulling. Red for third nulling. Green for fourth nulling. A second nulling goes after the item. This code applies only to flubbed lists as a whole — for instance whole list goes nul.

You can be left with two items in a list derived from a test item. Use both, but only if they are clearly of opposite character, not the same thing in another form.

At the end of nulling a test item list (first item of a line), you should have one or two live items. If one, put it under the line you’re doing on a Line Plot. If two, put one under the line you are doing and use the other for a new line. There are rarely two left on opposition lists.

When the item is found, check it out.

Get ruds in, run a bit of havingness.

See if item is still registering. If not get the ruds in better and do so until item reads well.

Now read an already nulled item on the list, then read the found item, then read a nulled item, then the found item.

Do this until you are sure all items on the list except the found item are nul.

If found item goes out, get the ruds in.

When you have found the item and checked it out, put it under its proper Line on the Pc’s Line Plot.

The Line Plot is a sheet of white foolscap (legal) with three columns across the top of each side, Line A, Line B, etc, with an indication of how each line was derived (Dislike, Like, Who O/W, Dynamic Assessment, etc).

Every one of these lines is itself. It does not cross over to other lines.

A Line is a list of found 3D items each in opposition to the last item on that Line. The Line is a series of zig-zags, with an item at each zig and at each zag. Any pair, a zig plus a zag, could be a 3D package that would run. We want at least five lines. We want all the items we can get on one line.

Inevitably, sooner or later, all lines will either coincide into a 3D package that will only derive itself when listed or the pc goes to OT by assessment.

There is a basic problem between every pair of items on one line in a Line Plot. Getting the pc to describe that problem helps blow charge.

When listing, differentiating or nulling, every time the pc gets a pain, write “PN” after the item. Every time an item makes a pc feel dizzy or he gets winds of space, write “SEN” after that item. When you finally come to run a package you could tell what is the pc’s term (pain) and what is the pc’s oppterm (sen) by studying the lists to see what type of item consistently gives the pc pain or sensation. Thus no error is made on selecting the terminal or further test needed.

The whole action 1 to 5 above is called Assessment.

The first error is poor E-Meter skill.

The second error is just lousy, ARC breaky auditing.

The third error is carrying a line by oppterms too deep beyond the other lines. Do lines one at a time in rotation. Don’t keep oppterming a line on and on and forget the other lines.

Fourth error is failing to note the ruds going out and getting off session invalidations and withholds.

Fifth error is not getting a long enough list to include the 3D item you’re after.

You can unburden a case of hundreds of found 3D items (thousands of list items). This makes terrific case gains, item by item found. You have never seen such fast case gains as a well done 3D Criss Cross by assessment alone providing the auditing is well done and these steps are followed.

Use only a Mark IV E-Meter. The others don’t register well enough to detect 3D Criss Cross reads.

Chanting a Modifier is not done in 3D Criss Cross.

Don’t let anybody not a Class II even attempt to learn 3D Criss Cross.